a synthetic study of the tropane and homotropane ring · pdf filebao gong teng (8) is an...
TRANSCRIPT
A Synthetic Study of the Tropane and
Homotropane Ring Systems
David E. Justice
A thesis submitted for the
Degree o f Doctor o f Philosophy in the
Faculty o f Science at the
University o f Leicester.
The Department of Chemistry The UniversityLeicester, LEI 7RH. MAY 1995
UMI Number: U070313
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a com plete manuscript and there are missing pages, th ese will be noted. Also, if material had to be removed,
a note will indicate the deletion.
Disscrrlâtiûn Publishing
UMI U070313Published by ProQuest LLC 2015. Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC 789 East Eisenhower Parkway
P.O. Box 1346 Ann Arbor, Ml 48106-1346
STATEMENT
The accompanying thesis submitted for the degree of Ph.D entitled "A Synthetic
Study of the Tropane and Homotropane Ring Systems" is based on work conducted
by the author in the Department of Chemistry at the University of Leicester mainly
during the period between September 1991 and October 1994.
All the work recorded in this thesis is original unless otherwise acknowledged in the
text or by references.
None of the work has been submitted for another degree in this or any other
University.
Signed: , ................................. .(iW. . . .
ACKNOWLEDGEMENTS
First and foremost I would like to thank John Malpass for all his help, support and encouragement over the past three years.
I would also like to thank fellow postgraduates and staff in the department who have helped me with my PhD. Special thanks go to Gerald Griffiths for NMR spectra, Graham Eaton for mass spectra, John Fawcett for X-Ray crystallography and Mick Lee for his expert technical advice.
Thanks go to Anne Crane for drawing the diagrams in this thesis and Claire for the proof reading.
Thanks also go to all the friends I’ve known at Leicester who have made my stay a memorable one - Ian Mason, Lee Carter, Mike Coogan and Craig Smith to name just a few.
Finally, I would like to thank my parents for their great support and optimism during the course of my studies.
ABSTRACT
A SYNTHETIC STUDY OF THE TROPANE AND HOMOTROPANE RING SYSTEMS, BY DAVID JUSTICE
1,4-Functionalisation of cyclohepta-1,3-diene using a nitroso-cycloaddition strategy provided precursors which were converted into the N-methyl-8-azabicyclo- [3.2.1]octane (tropane) ring system. Cycloocta-1,3-diene was used as the starting material to prepare precursors to the N-methyl-9-azabicyclo[4.2.1]nonane (homotropane) ring system.
Homotropane has been constracted, either with or without a bridgehead substituent, using mercury-mediated cyclisation of nitrogen onto an sp^ carbon centre. The versatile N-alkoxycarbonyl protecting group was employed to synthesise the coiTesponding norhomotropanes. This strategy was also used to prepare the 1-methyl- homotrop-7-ene analogue which could be epoxidised stereoselectively to yield the exo-epoxy derivative.
An efficient synthesis of homotropan-l-ol and norhomotropan-l-ol is described. Using variable-temperature ^H and ^ C NMR spectroscopy these bicyclic hemi- aminals were shown to exist in tautomeric equilibria with monocyclic amino-ketones. A range of derivatives were made; it was found that the incorporation of a double bond or an epoxide group into the 2-carbon bridge of the homotropan-l-ol structure altered the position of equilibrium. Difficulties were encountered in making quantitative measurements for some tautomer ratios and in these cases qualitative estimates were made. An identical synthetic strategy was used to make the homologous tropan-l-ols. This work includes a five-step synthesis of the tropane alkaloid physoperuvine (N-methyl-8-azabicyclo[3.2.1]octan-l-ol) in racemic form and in an overall yield of 79%. Norphysoperuvine (8-azabicyclo[3.2.1]octan-l-ol), unsaturated, and epoxide derivatives were also prepared and a study of their tautomerism with amino-ketones was investigated.
A steroselective method of introducing an epoxide group into the 2-caibon bridge of homotropane was devised. Homotrop-7-ene, protected with an N-alkoxycarbonyl group, was prepared by intramolecular displacement at an sp carbon centre by nitrogen. Epoxidation then gave the exo-epoxide but the overall yield was low because of difficulties associated with the cyclisation step. A more attractive procedure was developed which involved epoxidation of an allylic alcohol prior to cyclisation and provided access to both exo- and en^/o-epoxides in improved yields. Final removal of the N-alkoxycarbonyl protecting group at the end of the synthesis showed a substantial difference in epoxide stablity. The epoxide group of the endo-isomev was ring-opened by both hydride and hydrogenolysis reduction, whereas the exo-epoxide was suprisingly more resistant to attack. The homologous epoxides of the tropane series were prepared using a similar methodology. Again, the exo-epoxide was substantially more stable to reduction than the enc/o-epoxide.
Introduction of oxygen functionality at the C3 position of the tropane ring system was developed using cyclohepta-3,5-dienol in the initial nitroso-cycloaddition reaction. Combining this with the established approach to epoxy-tropanes led to the synthesis of the alkaloid scopine. The versatility of this procedure is demonstrated with the preparation of pseudoscopine and novel nor-derivatives.
CONTENTS
CHAPTER ONE INTRODUCTION PAGE No1.1 Tropane alkaloids 1
1.2 Established routes to hopanes 5
1.3 The nitroso cycloaddition / intramoleculardisplacement approach to tropanes 12
1.4 Homotropane alkaloids 18
1.5 Established routes to homotropanes 20
CHAPTER TWO THE TRANSANNULAR AMIDOMERCURATION STRATEGY TO HOMOTROPANES AND1-METHYLHOMOTROPANES
2.1 Introduction 27
2.2 Synthesis of 1 -methylhomotropane and1 -methylnorhomotropane 31
2.3 Synthesis of homotropane and norhomotropane 37
2.4 Synthesis of 1 -methylhomotrop-7-ene 41
2.5 Epoxidation of l-methylhomotrop-7-ene 49
CHAPTER THREE SYNTHESIS AND TAUTOMERISM OF 9-AZABICYCLO[4.2.1]NONAN-l-OLS
3.1 Introduction 53
3.2 Synthesis of homophysopemvine,norhomophysoperuvine and dehydro-derivatives 56
3.3 An efficient synthesis of homotropan-l-ols 60
3.4 Synthesis of exo-7,8-epoxyhomotropan-l-ols 62
3.5 Conclusion 64
CHAPTER FOUR 8-AZABICYCLO[3.2.1]OCTAN-l-OLS; PAGE NoSYNTHESIS OF PHYSOPERUVINE, NORPHYSOPERUVINE &DEHYDRO-DERIVATIVES
4.1 Introduction 68
4.2 Synthesis of physoperavine and norphysoperuvine 72
4.3 Dehydro-derivatives of physoperuvine 75
4.4 Conclusion 78
CHAPTER FIVE STEREOSELECTIVE OXYGENATIONOF HOMOTROPANE
5.1 Introduction 81
5.2 Synthesis of exo-7,8-epoxyhomotropane 81
5.3 Synthesis of enJo-7,8-epoxyhomotropane 88
5.4 Reduction of N-protected epoxyhomotropanes 91
5.5 Conclusion 93
CHAPTER SIX STEREOSELECTIVE OXYGENATIONOF TROPANE
6.1 Introduction 94
6.2 Synthesis of exo-6.7-epoxytropane 94
6.3 Synthesis of enJo-6,7-epoxytropane 98
6.4 Reduction of N-protected epoxytropanes 102
CHAPTER SEVEN TOTAL SYNTHESIS OF SCOPINE,PSEUDOSCOPINE AND NOR-DERIVATIVES
7.1 Introduction 104
7.2 A strategy to scopine and pseudoscopine precursors 108
PAGE No7.3 Synthesis of scopine and norscopine 112
7.4 Synthesis of pseudoscopine and norpseudoscopine 114
7.5 Further functionalisation of tropane 117
APPENDIX ONE X-Ray crystal data for the epoxidecycloadduct (329) 120
CHAPTER EIGHT Experimental 126
REFERENCES 213
ABBREVIATIONS
b.p. boiling point
°C centigrade
cm'^ wavenumber
Decomp. decomposition
DEPT distortionless enhancement by polarisation transfer
DIB AH diisobutylaluminium hydride
DMF dimethylformamide
DMSG dimethylsulphoxide
hr hour
Hz hertz
IR infra-red
LAH lithium aluminium hydride
lit. literature
M+ molecular" ion
MHz megahertz
min minute
m.p. melting point
mbar millibar
mmol millimole
MCPBA m-chloroperoxybenzoic acid
MEM P-methoxyethoxymethoxy
NMR nuclear magnetic resonance
ppm parts per million
TLC thin layer chromatography
TBAF tetrabutylammonium fluoride
TBDMSCl r-butyldimethylsilyl chloride
THF tetrahydrofuran
CHAPTER 1
Introduction
1.1 TROPANE ALKALOIDS
The tropane alkaloids are found in the plant families solonaceae,
convolvulaceae, erythroxylaceae, euphorbiaceae, proteaceae and rhizophoraceae.
They are hydroxylated derivatives of tropane (8-methyl-8-azabicyclo[3.2.1]octane)
(1) or nortropane (2) (figure 1.1).
R=Me Tropane (1)R=H Nortropane (2)
7
3Figure 1.1
These alkaloids form a large class of natural products (approximately 200) with
new structures continually being discovered.* Their interesting pharmacological
properties and varied uses in medicine have prompted significant research into
discovery, biogenesis, stereochemistry and synthesis.^ This thesis describes the
synthesis of both natural and novel non-natural tropanes and also of higher
homologues, homotropanes.
Tropanes can be sub-classed according to the degree of hydroxylation, namely
mono, di, and trihydroxytropanes. Atropine (3) and (-)-hyoscyamine are both tropic
acid esters of monohydroxylated tropane, isolated in 1833 (figure 1.2). Atropine is
MeN
H
(±)-Hyoscyamine (atropine) (3)
Figure 1.2
HOCH2 H
the racemic form of (-)-hyoscyamine and derives its name from the plant source from
which it originates {atropa belladonna, more commonly known as deadly nightshade,
a solonaceous plant). Other plants in this family include henbane (hyoscyamus
niger), the source of hyoscyamine and the thorn apple (datura stramonium). Both
these alkaloids are of historical interest in medicine; hyoscyamine was refered to in
Eber’s papyrus (ca. 1550 B.C.) as a treatment for ‘magic of the belly’ and was used to
remedy abdominal distress. The name belladonna (beautiful lady) arose from the use
of the plant’s juices as a cosmetic drug to dilate the pupil of the eye. Extracts of
atropa belladonna are still used as an antispasmodic and sedative in the treatment of
gastrointestinal disorders.
Cocaine (4) was first isolated in 1862 from erythroxylon coca lam^ which is
indigenous to Peru and is a diester of tropan-3|3-ol-2p-carboxylic acid (figure 1.3).
The leaves of the coca plant have long been known as a central nervous system
stimulant and were chewed by South Americans to alleviate the sub-zero temperatures
of the regional climate. The isolation of cocaine led to its use as a local anaesthetic
but because of its addictive and unpredictable nature the uses today are limited to eye,
nose and throat anaesthesia. The recognition of the properties of cocaine has resulted
in considerable research towards the synthesis of non-addictive analogues such as
novocaine,^ which still retain anaesthetic properties.
MeN
OCH
Cocaine (4)
A , Ph
Figure 1.3
Scopolamine can be formally considered as being derived from the
trihydroxytropane skeleton. It is the tropic acid ester of scopine (6) and occurs as the
racemic form in datura meteloides and as the (-)-form, named (-)-hyoscine (5), from
hyoscyamus muticus. On acid or base hydrolysis,® (-)-hyoscine is saponified to
(-)-tropic acid and scopine (figure 1.4). Scopine reacts further under basic conditions
to afford scopoline (7).
MeN
MeN
Add or base
hydrolysis
HOPh
HOCHz H(-)-Hyoscine (5)
Ph
HOCH.
HO
Scopoline (7)
(-)-tropic acid
Scopolamine possesses sedative properties and is found in various sleep aids,
more recently being used as a medicine for nausea. Scopolamine is also effective as
an antidote for organophosphate nerve gases and was carried by troops in the recent
Gulf war. Scopine, including a synthesis, is discussed in more detail later (chapter 7).
Bao gong teng (8) is an example of a recently discovered tropane alkaloid
(figure 1.5) and was isolated from the Chinese herb ericibe obtusifolia, having the
trivial name 'benth'. Traditionally the herb was used as a medicine for the treatment
HN
Bao gong teng (8)
of fevers but unfortunately had associated side-effects.^ It has since been discovered
that bao gong teng is an effective medicine in the treatment of glaucoma.® The
OHAcO
Figure 1.5
structure of (8) was assigned by a group of Chinese researchers® in 1981 and the
absolute configuration determined^ in 1989. As a result of its pharmacological
properties, (8) has been the subject of two recent syntheses.*®
Although it is not a tropane derivative, the alkaloid epibatidine (9) is based on
the closely related 7-azabicyclo[2.2.1]heptane skeleton (figure 1.6) and is the first
member of a new class of alkaloids. It was first isolated in 1992;** it was extracted
from the poison frog epipedobates tricolor, used by Ecuadorian Indians to poison
arrow tips.
Epibatidine (9)
H
Figure 1.6
Daly** discovered that the analgesic potency of (9) was 200 to 500 times greater
than that of morphine, and interestingly did not appear to bind to opiate receptors.
The unusual pharmacological and novel structure of (9), coupled with the scarcity of
obtainable material from natural sources, started a race to produce epibatidine
synthetically. Broka*^ reported the first synthesis of (±)-(9) in 1993. This utilised an
intramolecular displacement strategy as the key step, based on the pioneering
synthesis of 7-azabicyclo[2.2.1]heptane by Fraser and Swingle.*^ Fletcher*^
independently published a synthesis of (±)-(9) and resolved an intermediate racemate
to afford (+)- and (-)-epibatidine. A second enantiomeric synthesis was reported by
Corey.*^ The rapid accumulation of syntheses*® of epibatidine exemplifies the
interest generated by azabicyclic compounds in natural product chemistry.
1.2 ESTABLISHED ROUTES TO TROPANES
The diverse range of tropane alkaloids has inspired numerous syntheses of the
8-azabicyclo[3.2.1]octane ring system. Efforts have focused in two main areas;
firstly, the search for novel methods for preparing the basic tropane system and
secondly, the application of these approaches to introduce functionality. The
accumulation of this research ultimately leads to the syntheses of alkaloids. Some
general synthetic strategies are described here with recent examples of specific
natural product preparations.
Robinson developed what has become a classical synthesis of tropinone (10)
in 1917. This involved a Mannich base reaction between succindialdehyde,
methylamine and acetone (figure 1.7) and forms the basis of the industrial preparation
today.
MeN
+ MeNH2 + O
Tropinone (10) qFigure 1.7
The yield was subsequently improved*^ by replacing acetone with the calcium
salt of acetone dicarboxylic acid and altering the reaction conditions (pH 11, 20°C).
Further derivation of the Robinson synthesis was made by Stoll who introduced a
hydroxyl group at the Q position (figure 1.8). Replacing succindialdehyde widi a
furan derivative led to the preparation of 6P-hydroxytropan-3-one (14). Reaction of
the dihydrofuran (11) with hypobromous acid, elimination and subsequent
hydrogenation gave the 3-hydroxy-2,5-diethoxy-tetrahydrofuran (12). Hydrolysis to
the dialdehyde (13) followed by reaction with methylamine hydrochloride and
acetone dicarboxylic acid gave (14) in good yield.
OEt OEt
O
OEt (11)
HOBr
HO
OEt
OEt
(l)KOH
HO
OEt (12)
Figure 1.8HOAtgO
MeN O
HO
(14)O
"Cl
o (13)
The Robinson strategy has been used to synthesise other azabicycles: for
example, the homologue of tropane ypelletierine^ (15) was prepared by replacing
succindialdehyde with glutaraldehyde (figure 1.9). Unfortunately, the Robinson
methodology could not be used to prepare epoxide derivatives and failed when
epoxysuccindialdehyde was used.^i This method is therefore not applicable to the
preparation of scopine.
Figure 1.9 V pelletierine (15)
Bottini^ produced tropinone (10) by condensing 2,6-cycloheptadienone (16)
with methanolic methylamine to give a 95% yield of (10). Other N-substituted
tropinones could also be prepared using this approach^ (figure 1.10). Replacing
methylamine with hydroxylamine afforded N-hydroxytropinone (17).
RN
+ RNH2
(16)Figure 1.10
R=Me, (10) R=OH, (17)
An earlier synthesis of tropinone by Willstatter and Pfannenstiel^ used a
Dieckmann ester condensation in the key step (figure 1.11). Heating the pyrrolidine
(18) gave the pyrrole (19) which was subsequently hydrogenated and cyclised.
Saponification and decarboxylation of 2-carbethoxytropinone (20) afforded (10).
COzEtO
o
MeNH2
C H œ 2Et
NMeHeat
Figure 1.11
MeN
CHC02Et
(18)
MeN
NMe
C02Et
[H]
(10)o
Dieckmann
(20)O
C02Et
NMe
C02Et
A variation of this method allows for the synthesis of non-natural analogues of
tropane^^ (figure 1.12). N-Tosylation of cis-2,5-dicarbethoxypyrrolidine (21)
followed by hydride reduction and chlorination gave cis-N-tosyl-2,5-bis-
(chloromethyl)pyrrolidine (22). Condensation of (22) with phenylacetonitrile and
sodium amide afforded (23) as the only stereoisomer. This was ultimately converted
to 3-phenyltropane-3-carboxylic acid (24).
COgEt
NH
COzEt
CO,Et
TsCl/pyridine ^NTs
Figure 1.12MeN
DUAIH4
2)5002
(24)Ph
steps
NTs
CHgO (22)
PhCH2CN/NaNH2
TsN
(23)Ph CN
Based on an earlier observation by Bapat,^ Tufariello^ designed a synthesis of
tropan-3p-ol (28) using a nitrone-induced cycloaddition reaction (figure 1.13).
4-Nitrobut-l-ene was treated with acrolein and sodium methoxide and acidification
O
+1) NaOMaMeOH
»2)HO(g)
Figure 1.13
C H (O M e)2 l)Zn/NH4Cl(aq)
2)H a(»q)
(26)
MeN
Heat/Toluene
N O
OH
DMcI2)UA1H4
(27)
H
with hydrogen chloride gave the nitroacetal (25). Reaction of (25) with zinc in
aqueous ammonium chloride and subsequent acidification yielded the intermediate
nitrone (26) which cyclised to the isoxazolidine (27) on heating. Quaternisation with
methyl iodide and reduction with lithium aluminium hydride afforded tropan-3p-ol
(28) in 4 steps.
The methods described so far have involved a limited range of functionality. In
order to synthesise tropane alkaloids such as scopolamine a means of elaborating the
2-carbon bridge is needed. Some methods are known: for instance, oxyallyl
intermediates have been used to prepare trop-6-enes. Turro and Edelson^
synthesised 2,2-dimethyltrop-6-en-3-one (30) by reacting 2,2-dimethylcyclo-
propanone (29) with N-methylpyrrole (figure 1.14).
MeN
MeMe
NMeO
Me
Me
(29)
Figure 1.14
A similar approach was applied by Hoffmann:^® dehalogenation of a , a -
dibromoketones in the presence of N-substituted pyrroles afforded functionalised
trop-6-enes (31), again via an oxyallyl intermediate. A disadvantage of this method is
that the reaction only proceeds in acceptable yield when the a ,a ’-dibromoketone is
highly substituted (figure 1.15).
MeN
NMeM e - r
MeMe
Me'Figure 1.15 (31)
Noyori^ used oxyallyls generated from 1,1,3,3-tetrabromoacetones and
nonacarbonyldiiron. This allowed a reduction in the degree of substitution, but
involved the use of hazardous and expensive reagents. More recently, Mann^* has
reported a highly efficient and versatile method of preparing tropanes using oxyallyls.
This is discussed in chapter 7.
Modem preparative methods often target speciAc alkaloids, rather than general
approaches to the ring system. Davies^^ designed an elegant synthesis of
anhydroecognine (36) and ferruginine using vinyl carbenoids. Reaction of the
N-protected pyrrole (32) with a vinyldiazomethane (33) in the presence of
rhodium(II) hexanoate catalyst gave (34) in 75% yield. Selective hydrogenation and
deprotection with tetrabutylammonium fluoride afforded anhydroecognine methyl
ester (35). Reductive méthylation with formaldehyde and sodium cyanoborohydride
gave anhydroecognine (36) in excellent overall yield (figure 1.16).
C02MeNR N, RhgtOHex)
hexaneCOgMe
(32)
R=C02CH2CH2SiMe3
MeN
(34)
(36)
1) (PPhgyRhd/Hg2)Bu.NF
HN
H2CO/Na(CN)BH3
Figure 1.16
Pandey^^ used cyclic azomethine ylides in [3+2] cycloaddition reactions, the
ylide being generated by a sequential double desilylation method (figure 1.17).
(35)
10
Reaction of the precursor (37) with silver fluoride and phenylvinylsulphone in situ
gave the tropinone derivative (39) via the ylide (38). Desulphonlyation was achieved
with Raney nickel to afford tropinone (10) in good yield.
O
SiMe^
Me
(37)
MeN
PhSO;
o
Figure 1.17
Raney Ni
The versatility of this azomethine ylide approach was demonstrated by
Pandey*^ with a synthesis of epibatidine (figure 1.18). Using analogous precursors
to the tropinone synthesis, a [3+2] cycloaddition reaction between the disilyl-
SiMe,
CHjPhN
NCHzPh
SiMeg
(40)
Cl
(42)
Cl
(41)
Figure 1.18
11
pyrrolidine (40) and the cinnamate ester (41) gave the cycloadduct (42) in 83% yield.
Radical decarboxylation and débenzylation resulted in an efficient synthesis of
epibatidine (9).
1.3 THE NITROSO CYCLOADDITION / INTRAMOLECULAR
DISPLACEMENT APPROACH TO TROPANES
In 1984 Kibayashi ' reported a new approach to tropanes, the key step involving
a Diels-Alder cycloaddition of a cyclic diene with a nitroso compound. This original
work has since been considerably developed and has led to the synthesis of various
natural products and homotropanes. The first target Kibayashi prepared was
N-benzoylnortropane (figure 1.19). Reaction of cyclohepta-1,3-diene (43) with an
acyl nitroso reagent, generated in situ by oxidation of benzohydroxamic acid, afforded
the cycloadduct (44) in high yield. Reduction of the N-O bond with sodium amalgam
gave (45) which was hydrogenated to the 1,4-functionalised cycloheptane (46). A
portion of (46) was mesylated, and the remainder was treated with thionyl chloride
and triethylamine to yield the tran^-l,4-amido chloride (47). Attempts to cyclise the
cw-l,4-amido mesylate (48) using a range of strong bases failed. However, the
rrans-chloride (47) was cyclised to N-benzoylnortropane (49) in 87% yield using
potassium r-butoxide in hexamethylphosphoramide (HMPA) and benzene solvent.
This strongly suggests that an S^2 process is operating, with a trans-\,A
stereoelectronic arrangement of nucleophile and leaving group required for
cyclisation.
Attempts to hydrolyse (49) to nortropane failed, hence tropane (2) could not be
prepared via a benzoyl substituted nitrogen. Kibayashi tried to introduce different
protecting groups at nitrogen, namely carbamate groups. This created difficulties in
the chlorination step, as an elimination mechanism competed, resulting in the
formation of olefinic products. The cyclisation of such olefins is discussed in detail
later (chapter 2).
12
(43)
COPh
87*
,COPhHNCOPh
Namg77*
OH(44)
HNCOPh
(45)
HNCOPh
(47)
MsO/EtaN
Hypd/c84*
^ HNCOPh
Cl Ms=MeS02(48)
OMs
KO'Bu/HMPA/CfiHg87*
COPhN
Nortropane (2)Figure 1.19
Kibayashi^ demonstrated the practical uses of this approach, despite the
difficulties encountered, by synthesising tropane, tropan-3P-ol (28) and tropacocaine
(50) (figure 1.20). The latter two were prepared by elaborating the diene used in the
13
cycloaddition reaction.
MeN
Tropacocaine (50)Figure 1.20
Fraser and Swingle*^ had previously reported an intramolecular displacement
strategy for preparing 7-azabicyclo[2.2.1]heptane (52). The cyclisation step involved
treating the primary amine (51) witii sodium hydroxide in aqueous ethanol (figure
1.21). This is in contrast to the Kibayashi procedure of cyclising an amide-nitrogen
NHcCl HN
NaOH/BOH/HgO
(52)Figure 1.21
with a strong organic base. The increased nucleophilicity of the amine nitrogen offers
a more attractive means of synthesising azabicycles as milder conditions can be
employed. Indeed Broka^^ has recently synthesised epibatidine, using the Fraser and
Swingle procedure, with the cyclisation step requiring no external base and
proceeding in high yield. A second epibatidine synthesis used a very similar
method.
Bathgate^^ developed the Kibayashi method by modifying the nitrogen
substituent prior to nucleophilic displacement (figure 1.22). A benzoyl group offered
a high yielding cycloaddition reaction and facile reduction to a benzyl group thus
generating a nucleophilic nitrogen for cyclisation. Such a substituent is also
potentially removable at the end of the syndiesis. The cycloadduct (44) was prepared
from cyclohepta-1,3-diene (43) using a very similar procedure to Kibayashi.^ The
14
COPh
(44)
HNCOPh
Aimg92*
+ C1- HgNCHzPh HNCHoPh
Pyndine
soaycHag UAIH4
99*
Hg/Pd/C99*
HNCOPh
HO
(S3)
HO
(46)
Figure 1.22
CHzPhN
HN
Nortropane (2)HgM/C
95*
N-O bond was reductively cleaved and then hydrogenation of the olefin (45) afforded
the CM-amido alcohol (46) which was reduced to the amine (53) using lithium
aluminium hydride. Chlorination of (53) with thionyl chloride proceeded smoothly to
give the frons-1,4-chloro-amine (54) as the hydrochloride salt. No external base was
required for this step as the amine nitrogen acted as an internal base, reacting with the
hydrogen chloride formed and generating the required inversion of stereochemistry.
Direct basification of the crude hydrochloride (54) with pyridine gave the free amine
which readily cyclised to N-benzylnortropane (55). Hydrogenolysis reduced the
15
benzyl group giving nortropane (2).
Bathgate^^ then applied this approach to preparation of the unsaturated
analogues (figure 1.23). Reduction of (45) gave the amine (56) which reacted with
thionyl chloride and lithium chloride to give the tranj-1,4-chloro-amine salt (57).
HNCOPh
HO
97%
HO
+ Cl- HzNCHgPh
s o a j /L ic i
Cl(45)
Figure 1.23
(56)
CHzPhN
(57)
^ KgCOytJhrmsound
65%(59)
+
10%
Addition of the heterogeneous base potassium carbonate to the crude hydrochloride
(57) in an ultrasound bath resulted in cyclisation to benzylnortrop-6-ene (59) in 65%
yield. This was accompanied by an aziridine by-product (58), formed by 1,2 rather
than 1,4-cyclisation. Attempts to debenzylate (59) were unsuccessful. This method is
nonetheless a significant entry into tropane synthesis, being a high yielding
preparation of the uncommon trop-6-ene system.
At the same time an elegant method of preparing tropanes was published by
Bëckvall,^ using a similar approach to Bathgate. Palladium-catalysed single step
1,4-chloroacetoxylation of cyclohepta-1,3-diene (43) generated a 1,4-functionalised
cycloheptane and is illustrated here by the synthesis of a simple tropane derivative.
Stereoselective cw-chloroacetoxylation of (43) and substitution of the chloroacetate
(60) with sodium p-toluenesulphonanüde afforded the protected amide (61).
Hydrogenation, saponification of the acetate and mesylation gave the trans-
16
1,4-mesylate (62). Cyclisation was achieved using potassium carbonate as base in
metiianol solvent (figure 1.24). Deprotection of (63) was not reported, but Backvall
used this approach to synthesise 3-oxygenated stuctures, such as pseudotropine.
Attempts to cyclise unsaturated substrates failed. Nonetheless, this is a versatile
procedure that has been elaborated to prepare more complex tropanes such as scopine
and derivatives^^ (Chapter 7).
Cl
Pd(OAc)2/Ua/UOAc/Benzoquinone, 71%
HNTs
NmNHT:
70*
AcO(60)
AcO
Figure 1.24HNTs
KgCOgAleOH
77*
MsO(63) (62)
17
1.4 HOMOTROPANE ALKALOIDS
The second ring system discussed in this thesis is the higher homologue of
tropane, the 9-azabicyclo[4.2.1]nonane system. The N-methyl and secondary amino-
derivatives are called homotropane (64) and norhomotropane (65) respectively (figure
1.25).
R=Me Homotropane (64) R=H Norhomotropane (65)
Figure 1.25
In contrast to the large array of alkaloids based on tropane, homotropanes are
much less common in nature. Only two homotropane alkaloids have been discovered
to date, both based on norhomotropane, namely anatoxin-a^* (66) and homoanatoxin^^
(67). The former was isolated from blooms of toxic freshwater algae, anabaena
flos-aquae. Ingestion by wildlife of water containing high concentrations of this algae
has resulted in fatal poisoning. Appropriately, it was named ‘very fast death factor’
(VFDF) and, after stractural elucidation by X-ray crystallography*® and
spectroscopy,'** was renamed anatoxin-a (figure 1.26).
HN
° R = Me Anatoxin-a (VFDF) (66)^ R = Et Homoanatoxin (67)
Figure 1.26
Biological studies have shown anatoxin-a to be a potent muscarinic and
nicotinic agonist,'* and this has engendered considerable interest by organic chemists
in designing syntheses. Campbell'*^ reported the first synthesis of optically active
(66) via a ring expansion of a cocaine derivative. Rapaport^ used iminium salts to
construct (66) and this method was later developed into a chirospecific synthesis'*^
18
using D- and L-glutamic acids. Tufariello'^ modified the previously described
nitrone approach to tropanes^ and reported a preparation of racemic material.
Another synthesis was reported by Danheiser/^ the key step involving a transannular
substitution procedure. Although other approaches are k n o w n , a recent example is
discussed here: Somfai^^ reported a short, enantioselective method based on the
cyclisation of an N-tosyl iminium ion (figure 1.27). Starting with a DIB AH reduction
OTBDPS OTBDPS OH
NTsDIB AH. 92%
NTsSteps, 84%
NTs
(70)OMe
TBDPS=*BuPh,SiSteps, 76%
(l)HOA(c(m
(2) DBU/A
Steps, 69%NTs NTs
OMe OMeNa/Hg
▼Anatoxin-a (66)
(72) (71)
Figure 1.27
of the N-tosyl lactam (68) (derived from L-pyroglutamic acid^°), the a-hydroxy
sulphonamide (69) was prepared as an inseparable mixture of isomers. This was
converted to the a-methoxy sulphonamide which, after removal of the silyl protecting
group could be recrystallised as a single 2,5-substituted pyrrolidine derivative (70)
(stereochemistry arbitrarily assigned). Deprotonation of (70) with n-butyllithium,
followed by addition of phenyl chlorothionoformate yielded the thionocarbonate ester
which was immediately reacted with allyltributyltin giving the alkene (71).
Ozonolysis and subsequent Wadsworth-Emmons reaction with dimethyl acetyl-
19
methylphosphonate afforded the (72). Acidification of (72) and elimination of the
intermediate chloride with base, effected the desired elimination giving the N-tosyl
derivative (73) which was deprotected to give anatoxin-a using sodium amalgam.
1.5 ESTABLISHED ROUTES TO HOMOTROPANES
Interest in the homotropane skeleton has largely been confined to anatoxin-a
with over ten syntheses reported to date. "^ General methods of preparing other
derivatives has for the most part, been relatively neglected. Early work by Cope^^ led
to the first synthesis of homotropane based on a Tiffeneau expansion of tropinone
(figure 1.28). The cyanohydrin (74) of tropinone (10) was hydrogenated to give the
primary amine (75) which was treated with nitrous acid and gave the ring-expanded
homotropinone (76) in 45% yield. Raney nickel reduction, elimination of water from
the alcohol (77) and a second hydrogenation yielded homotropane (64). Although this
MeN
MeN
HCN
(75)
HOCN
MeN
MeN
Homotropane (64)Figure 1.28
IHONO, 79*
MeN
OH
70*64*
(77) (76)
20
is an antecedent synthesis, the potential to prepare derivatives is somewhat limited.
Later work by Anastassiou^^ involved reætion of cyclooctatetrene (78) with
cyanonitrene, generated by thermal fragmentation of cyanogen azide. A dilute
solution of (78) in ethyl acetate was heated with cyanogen azide to 78°C and resulted
in the evolution of nitrogen and the formation of two products in 31% yield (figure
1.29).
CNN
/ = = \NCN
(80), 10%(78) (79), 21%
Figure 1.29
N-Cyano-9-azabicyclo[4.2.1]nona-2,4,7-triene (80) was formed by 1,4-addition
of cyanonitrene (figure 1.30). Repeat reactions at lower temperatures afforded (79)
exclusively, and under the optimum conditions for the formation of (80) the yield did
not exceed 10%. Hydrogenation of (80) using a rhodium catalyst reduced all three
double bonds, giving N-cyanonorhomotropane (81). More gentle hydrogenation^^
CNN
CNN
(80)Rh/H2
RIVH
Figure 130
CNN
HN
KOH/MeOH
(82) (83)
21
resulted in partial reduction giving the single dihydro-derivative (82). Hydrolysis
furnished norhomotrop-2,4-diene (83).
Hobson^ cyclised N-chloroamines as a means of preparing the homotropane
skeleton (figure 1.31). Thermolysis of ethyl azidoformate in an excess of
cycloocta-1,5-diene (84) gave (85) in high yield. Reduction of (85) with lithium
aluminium hydride afforded N-methylcyclooct-4-eneamine (86) and chlorination at
nitrogen was achieved with N-chlorosuccinimide (NCS). Addition of the radical
initiator azobisisobutyronitrile (AIBN) to a solution of (87) gave a mixture of
azabicycles (88) and (89), the 2-chlorohomotropane (88) being formed in 25% yield.
HNMe
/ = \
NCOgEt
EtOCONg UAIH4
(84) (85)
MeN
(86)
ClNMe
NCS
AIBN
(88)25% (87)Figure 1.31
Haufe^^ also used cycloocta-1,5-diene (84) as a starting material in electrophilic
haloamidation reactions to prepare homotropanes. This approach suffered the same
drawback as that of Hobson, with two isomeric azabicyclic systems being formed in
the cyclisation step. In 1979 Tardella^® synthesised the N-ethoxycarbonyl derivative
of norhomotropane, the pivotal step being cyclopropyl ring fission of
bicyclo[5.1.0]octan-2-one (90) with pyridinium hydrochloride. The mixture of
%-chloroketones (91) and (92) was converted into oximes and reduced with lithium
aluminium hydride. The amines were converted to N-ethoxycarbonyl derivatives and
22
separated by preparative gas chromatography. The homotropane (94) was isolated in
19% yield (figure 1.32). Again the formation of a second isomer, N-ethoxycarbonyl-
7-azabicyclo[4.2.1]nonane (93), reduced the efficiency of this strategy as a means of
preparing homotropanes.
Py.HCl
y
(90) (91), 35%
C02EtN
(DNHgOH(2)UAIH4(3)ElOCOa
(92), 41%
COgEt
Figure 132
(94), 19%
+
(93), 10%
A synthesis developed by Barluenga^^ used the conjugated cycloocta-1,3-diene
(95) as starting material. It has advantages over some of the previous methods
described, giving the 9-azabicyclo[4.2.1]non-7-ene structure (98) as the sole product
although it is only applicable to N-aryl derivatives. Heating (95) with aniline,
mercury(II) oxide and tetrafluoroboric acid yielded (98) in a one-step process. The
mechanism was envisaged as proceeding through the intermediate 1,4 adducts
(96a,b), with cyclisation occurring either directly by displæement of mercury by
nitrogen or via initial elimination of mercury to the allylic intermediate (97) with
subsequent cyclisation (figure 1.33).
23
HNPh HNPh
PhNH24Ig(VHBF4
DMF/MfC
(96b)
f
HNPh
(97)(98), 45% Figure 1J3
The shortage of versatile routes to homotropanes led Smith^^ to investigate the
potential derivation of the Kibayashi^ and Bathgate^^ nitroso cycloaddition strategy,
from tropanes to the higher homologues (figure 1.34). By increasing the ring size
COPh COPh CHzPhN
Steps
(100)> <
H CHgPhN
HN
Norhomotropane (65)Figure 1.34
24
from cyclohepta-1,3-diene (43) to cycloocta-1,3-diene (95) a simple approach to
homotropanes and homotrop-7-enes was developed. Thus, reaction of (95) with the
acyl nitroso compound generated in an identical manner to Bathgate^^ gave the
homologous cycloadduct (99). Cleavage of the N-O bond with aluminium amalgam
and further steps including cyclisation using tetramethylpiperidine resulted in the
preparation of both unsaturated (100) and saturated (101) N-benzylnorhomotropanes.
N-Benzyl norhomotropane (101) could be hydrogenolysed to norhomotropane
but this method was clearly not applicable to débenzylation of (100) as simultaneous
reduction of the double bond would occur. Attempts by Smith to debenzylate (100) to
(102) using alkali metals in ammonia or alkyl chloroformâtes failed. Bathgate^
experienced similar difficulties in attempts to debenzylate other azabicycles. Smith^®
therefore approached the synthesis of the N-methyl systems by protecting the nitrogen
with a benzyloxycarbonyl group which could be reduced to methyl at a later stage of
the preparation (figure 1.35). Cycloocta-1,3-diene (95) was reacted with benzyl
nitrosoformate, again generated in situ. Reduction of the cycloadduct (103) yielded
HN
Me
MeN
ACE-Cl
Norhomotrop-7-ene (102)
Steps
MeN
Homotropane (64)
Figure 1.35
25
the oxazine (104), which was transformed using essentially identical chemistry to that
described in the previous syntheses of N-benzylnorhomotropane. Both homotropane
(64) and homotrop-7-ene (105) were prepared; déméthylation of the latter using
a-chloroethyl chlorofonnate (ACE-Cl) proceeded cleanly to give norhomotrop-7-ene
(102).
This thesis describes firstly the synthesis of functionalised homotropanes using
intramolecular and amidomercuration strategies. The introduction of epoxide groups
into the homotropane system is examined in detail and this work is extended to
include epoxy tropanes, leading ultimately to the synthesis of scopine and its
derivatives. In addition a high yielding method of preparing homotropane and
tropane azabicycles with hydroxy groups at the exposition is discussed.
26
CHAPTER 2
The Transannular Amidomercuration Strategy to Homotropanes and 1-Methylhomotropanes
2.1 INTRODUCTION
The dibenzo[a,£f|cycloheptenimine MK-801® (106) has attracted considerable
interest as an anticonvulsant and neuroprotective agent (figure 2.1). The structure is
based on the 1-methylnortropane skeleton which is unknown in nature to date.
HN
MK-801(106)Me
Figure 2.1
In addition to MK-801, derivatives have also been prepared^ (figure 2.2).
Base-induced cyclisation of the intermediate (107) with potassium r-butoxide
followed by reductive cleavage of the N-methoxy group gave the homologous
dibenzonorhomotropane system (108).
(1) KO'Bu
(2)ZnmOAc
Me 1 (107)NHOMe
(108)Figure 2.2
The cyclisation of alkenyl amine systems via electrophilic initiation with
mercury (H) salts has been used to synthesise a variety of heterocycles®^ and natural
products.^^ Takacs^ employed this method to prepare N-acyl pyrrolidine and
piperidines from alkenyl amidals (figure 2.3). The amidal (109) underwent rapid
cyclisation mediated by mercury (H) acetate and mercury (H) trifluroacetate (1.5
equivalents of a 1:1 mixture) in acetonitrile. The intermediate organomercurial
27
chloride was reductively cleaved to afford a 1:12.1 cis.trms mixture of
diastereoisomers (110) in 90% yield.
(1) Hg(0Ac)2mg(02CCP3)2(2) UBH4 /THF
90*
(110) (109)Figure 2.3
Smith^® applied this procedure to synthesise the 1 -methylhomotropane ring
system (116) via cyclisation of an amide on to an sp^ carbon of an exo-methylene
group. Jones oxidation of the 1,4-amido-alcohol (111) gave the amido-ketone (112);
subsequent Wittig methylenation of the ketone gave (113), the precursor to
cyclisation (figure 2.4).
HNCOPhHNCOPh
Jones Reagent
93*
HNCOPh
Pb3P"CH2
89*
OH O CH2(1 " ) Flgure2.4
Stereoselective amidomercuration studies using 5- and e-alkenyl-carbamates led
Harding^ to conclude that mercury(II) acetate induces cyclisation under non
equilibrating (kinetic) control, while mercury(II) trifluroacetate induces cyclisation
under equilibrating (thermodynamic) control. Harding also demonstrated that the
addition of either acetate or trifluroacetate salts is non-stereoselective and the results
of Smith^® reflect this theory of kinetic and thermodynamic control. Reaction of
(113) with mercury (H) acetate resulted in two discrete non-interconvertible
trans-(114a) and ci5-(114b) acetoxymercurinium ions. Only the former has the
correct stereoelectronic requirements for cyclisation to the organomercurial (115).
Assuming equal formation of (114a) and (114b) the optimum yield for the reaction
would be 50%. However, if mercury (IT) trifluroacetate were to be used instead and if
28
thermodynamic replaced kinetic control, then the two mercurinium ions would be
interconvertible and the yield should be significantly improved (Figure 2.5).
HNCOPh
(113)
Figure 2.5
HNCOPh HNCOPhX=OAc
HgX HgX(114a) (114b)
COPh
HgX
N
NaBH^/THF
Me(116)
Treatment of (113) under the conditions used by Takacs^ led to the isolation of
(116) in 42% yield along with a hydroxylated by-product (117) in 22% yield (formed
by acylation and subsequent reduction of the acetate). The yield was raised to 93% by
omission of mercury(II) acetate, using 1.1 equivalents of mercury(II) trifluroacetate
only. The N-benzyl derivative (118) was prepared by lithium aluminium hydride
reduction of the amide (117) in high yield (figure 2.6).
29
HNCOPh HNCOPhCOPhN
(113)
HO
X=0 A(V02CCF3(1:1) 22%
XzOgCCFa 0%
Figure 2.6
Me
42%
93%CHgPhN
Me(118)
The amidomercuration procedure was then applied to the synthesis of the
corresponding unsaturated system (figure 2.7). Barium manganate, a milder oxidant
than Jones reagent, was used to convert the allylic alcohol (119) to an a,P~unsaturated
ketone which was converted into the exo-methylene precursor (120). Cyclisation witii
mercury (H) trifluroacetate and subsequent reduction with borohydride afforded (121).
HNCOPh
HO (119)
COPhHNCOPh
(1)BaMn04, 80% ^(2)Ph3R=CH2,79*
Me(120) (121)
Figure 2.7
This chapter extends the procedure developed by Smith. Variation of the
nitrogen protecting group extends the synthesis to the N-methyl system, in addition to
the nor-systems. The amidomercuration method is also used to prepare the parent
azabicyclo[4.2.1]nonane system without a bridgehead methyl group, i.e homotropane
(64) and norhomotropane (65).
30
2.2 SYNTHESIS OF 1-METHYLHOMOTROPANE AND
1-METHYLNORHOMOTROPANE
Smith used hydrogenolysis with palladium on charcoal in methanol to
debenzylate 1-methyl-N-benzylnorhomotropane (118) to 1 -methylnorhomotropane
(122) (figure 2.8).
CHoPh H.HC1N N
Hg/Pdma
97%
Me (118) (122)Figure 2.8
This procedure could not be applied to unsaturated benzyl-amines such as (59)
because of competitive hydrogenation of the double bond. Attempts by Bathgate^^ to
debenzylate (59) to the nor-derivative (123) using alkali metals in liquid ammonia or
alkyl chloroformâtes failed (figure 2.9). This problem was overcome in the synthesis
of homotropane (64) and homotrop-7-ene (105), described in chapter 1.5. An
N-benzyloxycarbonyl (R2N-CO2CH2H1) protected nitrogen (a carbamate) was used
instead of an N-benzoyl group (R2N-COH1) in the initial nitroso-cycloaddition and
through the synthesis; lithium aluminium hydride reduction ultimately afforded
N-methyl azabicycles.^
CH2Ph HN N
(59)Figure 2.9
It was decided to apply the amidomercuration method using the more versatile
carbamate group. Literature precedents existed for such cyclisations. Harding^^"
31
converted 3-[(benzyloxycarbonyl)amino]-5-hexene (124) into a tran5-2,5-dimethyl-
pyrrolidine (125) in excellent yield. Subsequent removal of the carbamate group
afforded the amine (126) as outlined in figure 2.10.
H,C
Hg(OAc)2/NaBHj
(124)
HN ^ 3 COgCHzPh
95*Me
AcOH/Ha
V
(125)Figure 2.10
90*N' Me M e’' x Me COjCHoPh H HCl
(126)
Danishefsky^^ prepared 5-coniceine in four steps (figure 2.11). Reaction of the
carbamate (127) with mercury(II) acetate afforded an organomercury intermediate
(128). Treatment of (128) with sodium trimethoxyborohydride and two mole
equivalents of methyl acrylate gave the methyl ester (129) in an overall yield of 64%.
Hydrogenolysis, cyclisation and then lithium aluminium hydride reduction of the
lactam resulted in a straightforward synthesis of 5-coniceine (130).
CH,Hg(OAc) 2
\ NCOzCHzPh
(127)
\ ^
(128)
NaB(OMc)3H64*
NCOzCHzPh
1 = ^C 02Me
N
5-coniceine (130)
(2)UA1H4
Figure 2.11(129)
The saturated 1-methylhomotropane system was the first synthetic target and an
attempt was made to prepare (136) having the benzyloxycarbonyl-protected nitrogen.
The previously described Diels-Alder^* reaction between cycloocta-1,3-diene and
32
benzylnitrosoformate generated in situ from benzyl-N-hydroxycarbamate and
tétraméthylammonium periodate afforded cycloadduct (103) in 61% yield and the
double bond was reduced to (131) using a diimide reduction.^ Disappointingly, the
yields did not exceed 40% in initial experiments using dichloromethane as solvent.
Hamersma^® reported that solvent had an important influence on reaction and indeed,
by using methanol, the yield of (131) was raised to 91%. The formation of (131) was
evident from the disappearance of the olefmic signals in the and NMR spectra,
some signals were broadened however because of restricted rotation about the
N-CO bond. The N-O bond of (131) was reductively cleaved in quantitative yield
using freshly prepared and powdered sodium amalgam,^^ buffered with sodium
phosphate (figure 2.12). Oxidation of (132) using the Jones procedure^ gave a high
(95)
61*
Figure 2.12
HNR
HN=NH/MeOH
91*
(103)
R=CO,CH,Ph
HNR
Jones rea^nt
98*
(133) (132)
yield of the ketone (133) isolated as a crystalline white solid. This showed a carbonyl
absorption at 1720 cm'* in the IR spectrum and a carbonyl signal at 5 217.0 in the
NMR spectrum. On leaving a solution of (133) to stand in CDCI3 for several hours a
second set of signals was detected. The *H NMR spectrum showed an AB quartet at 6
5.11 and 5 5.17 with a geminal coupling of 12.4 Hz. This corresponded to a
diastereotopic pair of protons of the benzyloxy CH2 group. A characteristic
33
quaternary signal at S 92.9 in the NMR spectrum was also present. These signals
were assigned to the tautomer of (133), the bicyclic hemiaminal (134), as depicted in
figure 2.13. The phenomenon of tautomeiism in 1,4-amino-ketone systems is
discussed in more detail in chapters 3 and 4.
HNCOzCHzPh COzCHzPhN
OH
Figure 2.13(133)
The alkene (135) was prepared using standard Wittig chemistry, following the
Corey procedure^ in which the methylene ylide is generated from methyl-
triphenylphosphonium bromide and dimsyl anion as the base (figure 2.14). This was
reported to be higher yielding than the classical Wittig method^ which used
n-butyllithium in diethyl ether to generate the ylide. The ketone was methylenated
accordingly in 48% yield. The formation of (135) was apparent on inspection of the
NMR spectrum; the carbonyl was replaced by a quaternary signal at 5 150.9 and a
new methylene signal was present at 5 111.6. This spectroscopic data was similar to
RN
R=C02CH2Ph
HNR HNR
48*
(136) 61%Me
N
(133) (135)
Figure 2.14
(137) 25%
OH
34
that of the closely related N-benzoyl derivative (113). Attempts to vary reaction
conditions or prepare the ylide from other bases and solvents made no improvement
to the yield. The intrinsic tautomerism process associated with (133) is the probable
cause of the modest yield. The exo-methylene compound (135) was then reacted with
1.1 equivalents of mercury(II) trifluoroacetate and reduced with sodium borohydride,
resulting in the isolation of (136) in 61% yield.
The and NMR analysis confirmed the structure of (136); although
complicated by the presence of a pair of rotamers, the NMR spectrum showed a
singlet corresponding to the methyl group at 5 1.60 (common to both rotamers). The
NMR spectrum included methyl signals at 5 28.6 (major rotamer) and 6 29.7
(minor rotamer). The desired product was accompanied by a by-product (137) in 25%
yield. The and NMR spectra of (137) differed to those of (136) with the
methyl signals being replaced with an AB quartet at 6 3.60 and 5 3.70 (J = 11.7 Hz) in
the NMR spectrum and a new methylene signal at 5 68.9 in the NMR
spectrum. The mass spectrum exhibited an increase in relative molecular mass of 16
from (136), indicating the incorporation of oxygen, and the 1-hydroxymethyl structure
was therefore assigned to (137).
The explanation for the formation of (137) is outlined in figure 2.15; reduction
of an organomercurial acetate (138) to (140) using sodium borohydride generates a
new C-H bond. This occurs via a free radical chain process, ' initiated by collapse of
the organomercurial hydride (139). Loss of elemental mercury then creates an alkyl
RHgRH
(140)
[R-Hg-H]R-Hg-OAcHg"
Figure 2.15
35
radical which abstracts hydrogen from (139).
Evidence for such a process is supported by deuterium aluminium hydride
studies.^^ Intermediates have also been trapped: Quirk^^ investigated the sodium
borohydride reduction of 2,2,2-triphenylethylmercury(II) chloride in both anaerobic
(equation 1) and aerobic (equation 2) conditions using basic aqueous THF as solvent
(figure 2.16).
Pb^CMe PhzCHCHgPh PbgQOHXmzPh Ph3CCH20HTW/H20
1. Ph3CCH2HgQ + NaBH ----------^ 8% 92* 0 * 0 *
THF/H2O2. Ph3CCH2HgCI + NaBH^ + O2 ^ 13* 3 * 58* 19*
Figure 2.16
The results indicate the formation of a radical proceeded by a 1,2-free radical
rearrangement. Such rearrangements are less common than 1,2-nucleophilic shifts
but are known for neophyl systems. '* The oxygen acts as a radical scavenger in
equation 2 and the oxygenated substrates are the predominant products. Further
studies by Whitesides^^ have added credence to this mechanism. This explains the
formation of the 1-hydroxymethyl azabicycle (137) as no inert atmosphere or
degassing methods were used. When the formation of (137) was suppressed using an
inert atmosphere, the yield of (136) increased to 73%.
Reduction of the carbamate (136) with lithium aluminium hydride afforded
N-methyl-l-methyl-9-azabicyclo[4.2.1]nonane (141) in 78% yield (figure 2.17). The
amine was volatile and was therefore handled and characterised as the hydrochloride
salt. The ^H NMR spectrum indicated a pair of diastereoisomeric quaternary
ammonium salts in an integrated ratio of 8:1. The major N-methyl signal was
observed at 8 2.73 as a doublet with a coupling of J = 4.4 Hz and was attributed to the
equatorial diastereoisomer by analogy with the work of Bottini.^® Hydrogenolysis
gave the secondary amine (122) which had identical spectroscopic properties to the
hydrochloride prepared by Smith.^*
36
M e H aN
Me
78* (141)
95*
H2/Pd4ia
H.HC1N
Figure 2.17(122)
2.3 SYNTHESIS OF HOMOTROPANE AND NORHOMOTROPANE
Kibayashi^ reported a new method of preparing tropanes using base-induced
cyclisations of 1,4-amido-chlorides (chapter 1.3). However, when N-benzyloxy-
carbonyl protected amino-alcohols were treated with thionyl chloride the formation of
dehydration products competed (figure 2.18). Thus, the benzyl carbamate (142),
when treated with 2.5 equivalents of thionyl chloride, afforded (143) in 12% yield
together with a 1:1 mixture of symmetrical and unsymmetrical cycloheptenes
(144a,b). In addition to base induced cyclisation of (143) using potassium r-butoxide
in benzene and HMPA, the cycloheptenes were also cyclised to (145) using the
amidomercuration method. The separation of (144a) and (144b) was not necessary as
both olefins afforded the required 8-azabicyclo[3.2.1]octane product (145). The 45%
yield was low, which can be assumed to be because the use of mercury (H) acetate
results in non-equilibrating conditions, the cis-mercurinium ion being unable to
cyclise.
A decision was reached to examine amidomercuration as a means of preparing
homotropane and norhomotropane. Using mercury (H) trifluroacetate instead of
37
HNR HNR HNR
a
HNR
+
(143), 12%
R=C02CH2PhK(/Bu
78% /(l)Hg(OAc)2(2)NaBH4,45%
RN
(145)
Figure 2.18
mercury (H) acetate was expected to result in a more efficient cyclisation than the 45%
yield recorded by Kibayashi in the tropane series. The cyclooctenes (147a,b) were
prepared from the ketone (133) using the Shapiro reaction^ (figure 2.19). The
tosylhydrazone was first synthesised using standard conditions/^ by refluxing (133)
with p-toluenesulphonyl hydrazide in ethanol with an acid catalyst. The *H NMR
spectrum was entirely consistent with the hydrazone (146) and confirmed the
incorporation of a tosyl group. The NMR spectrum was complex because of the
mixture of isomers present. Reaction of (146) with four equivalents of n-butyllithium
resulted in a pair of inseparable regioisomers (147a,b) in 42% yield, with olefinic
HNR HNR HNR HNR
(133) (146)
NHTsFigure 2.19
Buumtp
(147a)
RzCOoCHiPh
(147b)
38
signals visible in the NMR spectrum between Ô 5.57 and 5 5.80. Measurement of
the ratio (147a):(147b) from the NMR spectrum was difficult as all the signals
were superimposed but the ratio was estimated to be approximately 1:1 from
NMR integrations. The NMR spectrum showed two (CH) methine signals at 8
51.2 and 8 51.6 corresponding to the a-nitrogen carbon, and a total of nine (CHg)
methylene ring carbons (two overlapping).
The mixture of cyclooctenes (147a,b) was then treated with mercuric(II)
trifluroacetate and sodium borohydride, using identical conditions to the previous
cyclisation of (135). It was anticipated that cyclisation of the cyclooct-4-ene (147b)
to give the azabicyclo[3.3.1]nonane system (149) might compete. Fortunately this
was not the case and only the desired N-(benzyloxycarbonyl)-9-azabicyclo[4.2.1]-
nonane (148) was isolated in 73% yield (figure 2.20). Recognition of (148) was
HNR
R N
+HNR
(147b)
(148)
Figure 2.W
' (149)
straightforward because of the simplicity of the NMR spectrum. Restricted
rotation about the N-CO bond resulted in a pair of rotamers; three pairs of (CH2)
methylene signals were displayed at 8 24.1, 8 24.2, 8 29.3, 8 29.7 and 8 30.2, 8 31.3.
The [3.3.1] system (149) would have shown only two such pairs because of the higher
affiliated symmetry. The isolation of only (148) is in agreement with studies by
39
Barluenga^^ on the amidomercuration of cycloocta-1,5-diene (84). Using mercury(II)
acetate the [4.2.1] regioisomer was formed initially (under conditions of kinetic
control) but subsequent pyrolysis produced the [3.3.1] isomer as thermodynamic
replaced kinetic control.
The versatility of the carbamate group was then utilised. Lithium aluminium
hydride reduction of (148) afforded homotropane (64) in 99% yield. Hydrogenolysis
with palladium on charcoal gave norhomotropane (65) in 89% yield (figure 2.21).
Both amines were stored as hydrochloride salts and had identical spectral properties
to previously reported data. ®
Me.HClN
COiCHoPh UAIH /HQ
Homotropane(64)
H.HC1N
Norhomotropane(65)
Figure 2.21
A concise synthesis of homotropane and norhomotropane has been
demonstrated, the overall efficiency being reduced, however, by the disappointing
yield of the Shapiro reaction. Preparation of the cyclooctenes (147a,b) by other
means, such as dehydration of the alcohol (132) was not attempted although such an
approach would be expected to improve the yield significantly.
40
2.4 SYNTHESIS OF l-METHYLHOMOTROP-7-ENE
Attention was next turned to the synthesis of the l-methylhomotrop-7-ene
system (155), starting from the cycloadduct (103) but excluding the diimide step, as
outlined if figure 2.22. Reduction of (103) using sodium amalgam gave the allylic
alcohol (150) in 98% yield; the structure of (150) was confirmed by an exchangeable
proton at S 2.62 in the NMR spectrum. The presence of the double bond in (150)
required a milder method of oxidation than Jones reagent and manganese dioxide®®
was used initially affording (151) and (152) in 62% yield. Riouzabadi®' reported that
barium manganate was a preferable reagent for the oxidation of allylic alcohols and
an improvement to 85% yield resulted from its use.
HNR
Na/Hg
98%
(151) * OH
R=C02CH;PhHNR
Figure 2.22(153)
In accordance with the earlier observations concerning the unsaturated analogue
(133), an equilibrium between the monocyclic a,P-unsaturated ketone (151) and the
bicyclic hemiaminal (152) was shown to exist. Interestingly the interconversion
between the two was slow and separation was possible using chromotography. The
a,P-unsaturated ketone (151) was identified by a characteristic carbonyl signal at 8
203.4 in the NMR spectrum and an IR absorbance of 1715 cm’^ The spectra of
41
the bicyclic tautomer (152) were more complex because of the presence of a pair of
rotamers. The NMR of the major rotamer showed two olefinic signals, as part of
an ABX system, with the Hg proton at S 5.74 having a vicinal coupling of 6.2 Hz to
H7 and a further w-coupling of 1.0 Hz. The olefinic proton H7 at 5 5.86, displayed a
further vicinal coupling to the H^-bridgehead proton of 2.6 Hz. The NMR
showed a quaternary (COH) signal at 5 95.7. The IR absorption of 1715 cm'^ was
absent for (151), but a broad 0-H stretch was visible at 3470 cm '\ If a solution of
either (151) or (152) in CDCI3 was left to stand, equilibration to a weighted ratio of
37:63 for (151) to (152) occurred over a period of days. Although such tautomerism
processes are discussed later, it is worth comparing (133) and (151) and noting that
the inclusion of a double bond shifts the equilibrium in favour of the bicyclic
tautomer.
Wittig methylenation of the a,P-unsaturated ketone proved troublesome. The
optimum conditions used a 5 molar excess of methylene ylide generated from dimsyl
anion and resulted in a 57% yield of (153). The NMR spectrum indicated the
absence of a low field carbonyl signal, which was replaced by a quaternary (C=CH2)
signal at 6 146.0 and a new methylene signal at ô 118.9. The ^H NMR spectram
showed the exo-methylene protons at S 4.87 as a broad singlet and at Ô 4.96 as a
doublet with a geminal coupling of 1.0 Hz. In repeat reactions the formation of (153)
was unpredictable, the yield varying between 40% and 0%. It was assumed that the
tautomerism process was responsible, although attempts to methylenate freshly
chromatographed a,p-unsatutated ketone (151) made no improvement.
As mercury(n) trifluroacetate had proved superior to mercui-y(II) acetate in
previous reactions, it was used in the attempted cyclisation of (153). This resulted in
unexpected failure; the only isolated product was an interesting hydroxylated
derivative (154) in 55% yield (figure 2.23). Resorting to mercury(II) acetate did
produce the required homotrop-7-ene (155) but, as anticipated, only in the modest
yield of 45%, for reasons discussed previously. In partial recompense for the
mediocre yield, some starting material was recovered. The major rotamer of (155)
42
RN
HO(i)Hg(02c a y )2
HNR Me53*
45*
RzCOzCHzPh
(1)Hg(OAc)2(2)NaBH4
(155)
Figure 2.23
displayed a methyl carbon signal at 6 24.3 and two alkene carbon signais at 8 127.8
and S 136.4 in the NMR spectrum. The ^H NMR spectrum displayed an ABX
system for the two olefinic protons H^, Hg and the bridgehead proton Hx Protons
H^ (S 5.49) and Hg (8 5.64) showed a mutual vicinal coupling of 6.2 Hz. In addition
Hg coupled vicinally (J = 2.6 Hz) to Hx, whereas H^ showed only had a small
w-coupling of 0.8 Hz. A corresponding set of signals in both the *H and NMR
were assigned to the minor rotamer of (155).
Extensive studies were made to elucidate the exact stereochemistry of (154).
The ^H and NMR spectra were complex at room temperature since they were
recorded below the coalescence temperature of the rotamers, caused by restricted
rotation about the N-CO bond. Attempts to assign signals to the two rotamers was not
possible with confidence, but notable features in the ^H NMR spectrum were an
exchangeable proton at 8 2.58 and a series of broad multiplets between 8 3.86 and 8
4.40 assigned to the Hg and a-hydroxy protons. Using variable temperature NMR and
heating (154) in Dg-toluene resulted in initial broadening followed by signal
coalescence at 363K. It became apparent that (154) was actually a pair of isomers, as
43
two individual bridgehead and two H7 or Hg a-hydroxy protons were identified. Up
to four isomeric structures (154a-d) were therefore possible (figure 2.24).
RN
4 #Hga
RN
(154a)
RN
HOMe
(154b)
R=C02CH2Ph RN
OH Me Me%OH
(154d)
Figure 2.24
Integration of the NMR signals indicated that the broad multiplet at 6 4.32 due
to the Hg-bridgehead was connected to the a-hydroxy doublet of doublets (J = 6.8,4.8
Hz) at 8 3.74. The signals at 8 4.14 and 8 3.64 were likewise related. This was also
confirmed by consecutive double irradiation experiments at 363K as illustrated in
figure 2.25. Irradiation of the Hg signal at 8 4.14 caused the broad (apparent) doublet
at 8 3.64 (J = 6.8 Hz) to simplify to a doublet of doublets with vicinal couplings of 1.9
Hz and 6.7 Hz. Irradiation of the Hg multiplet at 8 4.32 caused the doublet of doublets
at 8 3.74 to remain unchanged. Two points can be deduced from these irradiations;
firstly the doublet at 8 3.64 is vicinally coupled to the Hg proton (J < 1 Hz), and
therefore this proton must be either H ^ or H7R. Secondly as no perturbation was
observed when the Hg multiplet at 8 4.32 was irradiated, it can be reasoned that the
signal at 8 3.74 is either a Hgg or a Hgp proton, unless the vicinal coupling between
the two is effectively 0 Hz.
Some evidence is thus provided that (154) is a regioisomeric mixture. It was
44
300 MHz 'H NMR Double irradiation experiments
of the hvdroxvlated homotropane (154)
Figure 125
a-OH
a . 6 0 4 . 2 0 a . 0 0 3 . 3 0PPM
3 . 6 0 3 . 4 0
45
decided to oxidise a small portion of in an attempt to strengthen this proposal. If
regioisomers existed then two ketones would result, whereas stereoisomers would
afford only one such ketone. Oxidation of (154) using Jones reagent (figure 2.26)
resulted in a *H NMR spectrum which was complex at room temperature. The
oxidation was confirmed by a decrease in relative molecular mass of 2 with respect to
COzCH^Ph COoCHoPh COzCHgPhN N N
HgHO Jones reagent
Me(156a) (156b)
Me(154)
Figure 2,26
starting material. Variable temperature NMR spectroscopy was again used and
coalescence occurred at 363K. Inspection of the signals at elevated temperature did
indeed identify a mixture of 7-keto and 8-keto-1 -methylhomotropanes (156a,b).
These were recognised by the presence of two Hg-bridgehead protons. A broad
doublet at 6 4.02 (J = 8.4 Hz) was attributed to the 7-keto system (156a), vicinally
coupling to C5 protons. This was simpler than the corresponding Hg signal of (156b),
which showed additional coupling to the C7 protons, appearing as a broad multiplet at
5 4.32. These assignments were confirmed from double irradiation experiments (see
experimental).
It had now been demonstrated that (154) was indeed a mixture of regioisomers,
but the a- or ^-stereochemistry was still uncertain. This was rationalised after
norhomotropan-7P-ol (241) and nortiomotropan-7a-ol (263) had been synthesised
(chapter 5). Full spectral data for both were recorded and are tabulated below.
The vicinal couplings for the stereoisomer of (154) with the a-hydroxy proton
in either the H ^ or H^p position had values of J7 g = 6.7, 1.9 Hz and Jgj < IHz. The
other stereoisomer of (154) with the a-hydroxy proton in either the Hg^ or Hgp
position had values of J7 g = 6.8, 4.8 Hz and Jgj = 0 Hz. Comparison of these
46
H
H. ur»
Hi 3.30 He Hgfrs
He - 3.77 H?a
H?a — 4.12 Hga 1
Hga 8.7 — 6.2 1.97 Hgp
Hgp 3.3* — 2.8 14.5 1.88
♦Mutually assigned
\ 5H f \ Hi % l\Hi 3.46 He
He - 3.62 H?pT HHga
H?p - "7.0 4.40 HgaF
Hga 4.2* — 7.7 1.37 Hgp
Hgp "10* 10.0 13.4 2.43
'H ,
(263)
♦Mutually assigned
coupling constants with the corresponding values of norhomotropan-7P-ol (figure
2.27) and norhomotropm-7a-ol (figure 2.28) with (154a) and (154b) indicated a close
fit with norhomotropan-7P-ol (241) which displayed couplings of J7 g = 6.2, 2.8 Hz
and Jgj = 0 Hz. The respective values of norhomotropan-7a-ol (263) bore no
similarity to either (154a) or (154b), as all vicinal couplings were much larger. It was
therefore concluded with certainty that the mixture of alcohols was (154a,b) in which
both hydroxy groups were p. The isomer which displayed a small coupling of J5 7 < 1
Hz was assigned to (154a) and the isomer which displayed no such coupling was
47
assigned to the P-hydroxy isomer (154b).
Oxymercuration-demercuration*^ is a standard procedure for the Markovnikov
addition of water to an olefin. The mechanism proceeds by electrophilic addition of a
mercury (H) salt to a double bond and then sequential nucleophilic ring opening of the
mercurinium ion by water. Demercuration of the resultant organomercurial is then
effected by sodium borohydride reduction. It is postulated that the 7- and
8-hydroxy-1-methylhomotropanes (154a,b) were formed in such a manner, as a result
of a trace quantity of water in either the acetonitrile or THF solvent, prior to
reduction. The organomercurial (157) was presumed to have been formed and to have
undergone a second electrophilic addition with mercury(II) trifluroacetate, to give the
diorganomercurial species (158), which was subsequently ring opened by water to
furnish (159). The formation of (154a) is shown in figure 2.29.
HNR
HgtOzCCFglz
(153)
Figure 2.29 HO
RN
RN
Hg(02CCF3)2CFsCOzHg
----------
HgOzCCFg (igT) (158) HgO^CCFg
RN
CHq
(154a)
R = C02CH2PhHgO
RN
NaBH4 HO CFgCOzHg
(159) HgO^CCFg
Traylor^ demonstrated that for the strained carbobicycle norbomene,
oxymercuration-demercuration results in stereoselective exo-hydroxylation via the
formation of a cis-organomercurial intermediate such as (159) and this mechanism is
illustrated in figure 2.29. Although a more detailed mechanistic study would be
48
required to demonstrate the applicability of this to less strained systems such as the
homotrop-7-ene (157), the results of Traylor may well explain the observed
gjco-stereochemistry for the conversion of (153) into (154a,b).
A two molar stoichiometric quantity of mercury(II) trifluroacetate would be
required for the complete conversion of (153) into (154a,b). Only 1.1 equivalents
were used resulting in a 55% yield. Repetition of the experiment with a larger excess
of mercury salt, using standard aqueous oxymercuration conditions, was not
investigated because of the shortage of starting material. Nevertheless,
amidomercuration of (153) with mercury(II) acetate using rigorously dried solvents
afforded the desired product (155) without competitive oxymercuration products.
Lithium aluminium hydride reduction of (155) was not attempted as material was
required for epoxidation studies. In accordance with the reduction of the saturated
analogue (136), high yielding reductions would have been expected.
2.5 EPOXIDATION OF l-METHYLHOMOTROP-7-ENE
The 8-azabicyclo[3.2.1]octane system containing oxygen functionality in the Cg
and C7 locations (either as ester or hydroxyl groups) is frequently encountered in
nature. Scopolamine is a unique example of a tropane alkaloid containing a
6,7-epoxy group and the epoxide is a potential precursor of hydroxy- and
dihydroxy-compounds. The stereoselective introduction of such functionality is
therefore of key importance in synthesising natural products and their analogues.
Epoxidation of trop-6-ene derivatives offers a way of preparing epoxide
functionalised tropanes, and hence a diverse range of oxygenated derivatives.
Fodor® has reported a total synthesis of scopolamine, via the synthesis of
scopine (6), with the key step involving the epoxidation of 3a-acetoxytrop-6-ene
(160). Difficulties were associated with this reaction since competitive oxidation of
the amine resulted in N-oxide formation. Reaction of (160) with monoperoxyphthalic
acid afforded the N-oxide (161) as the major product. An excess of peroxyacid gave a
49
poor yield of O-acetyl-scopine N-oxide (162), which could not be reduced back to the
tertiary amine without simultaneous ring opening of the epoxide, affording the
alcohol (163) (figure 2.30).
MeN
o: Me o: Me'N+ 'N+
AcO AcO(161)
AcO(162)
MeN
Figure 2.30 HO
AcO
Epoxidation of (160) was achieved by first preparing the trifluroacetate salt of
(160) and then treating with trifluroperoxyacetic acid in dichloromethane giving
O-acetyl-scopine (164). This was subsequently saponified to scopine (6). No yield
was quoted for the epoxidation step and the reaction was slow, taking eight days for
completion (figure 2.31). Other syntheses of scopine are discussed in more detail in
chapter 7.
^MeN+
MeN
AcOAcOFigure 2.31
MeN
Acetone
HO
Scopine (6)
50
To assure efficient epoxidation a suitably protected nitrogen is therefore
required. Extensive studies by Koôovsky*^ reported high yielding epoxidations of
allylic and homoallylic carbamates. It was decided to attempt an epoxidation of the
N-alkoxycarbonyl-protected l-methylhomotrop-7-ene (155) to investigate the yield
and also test the belief that a pronounced exo-stereoselectivity would predominate, in
view of Fodor’s results. MCPBA (50 - 60% purity) was the chosen peroxyacid; when
(155) was stirred with 1.3 equivalents of MCPBA, (165) was isolated in excellent
yield as the only stereoisomer (figure 2.32).
COgCH^Ph COgCHzPhN N
MCPBA
94*
Figure 2.32
O
Me(165)
The epoxide (165) was identified as a pair of rotamers; the major rotamer
displayed characteristic epoxide proton signals in the NMR spectrum at 8 3.23 and
8 3.32 with a mutual vicinal coupling of 3.1 Hz, twing part of an ABX system,
including a small vicinal coupling (0.4 Hz) to the Hg-bridgehead proton at 8 4.36.
The NMR spectrum displayed two new methine signals at 8 56.5 and 8 57.6
replacing the olefinic signals of (155). The assignment of exo-stereochemistry was
based on three observations. Firetly epoxidation of norhomene®^ (166) resulted in the
exo-epoxide (167) as the sole product, with the approach of the peracid from the less
hindered face (figure 2.33). Model epoxidation studies by Howarth^ of N-protected
7-azabicyclo[2.2.1 jheptene systems such as (168) manifested the same
stereoselectivity, giving only the exo-epoxide (169). Secondly, on moving to larger
azabicycles, the exo-approach of peracids is preserved. Fodor only obtained the
exo-epoxide (164) on epoxidation of the trop-6-ene derivative (160). Thirdly in later
syntheses (chapter 5) both the exo- and endo-7,8-epoxyhomotropanes are prepared via
the epoxidation of allylic alcohols. The coupling constants and chemical shifts of
51
(166)
MCPBA
99*
(167)
CHgCOgPhN
MCPBA
86*
CHzCOgPhN
(169)
Figure 2 3 3
(165) in the NMR spectrum form a conclusive match with corresponding signals
of the N-protected cjco-7,8-epoxyhomotropane (235), prepared later.
Attempts to remove the carbamate group of (165) were not made because of the
shortage of material caused by the low yielding Wittig methylenation. Nonetheless,
subsequent deprotection of the gjco-7,8-epoxyhomotropane (235) was a
straightforward process (chapter 5).
52
CHAPTERS
Synthesis & Tautomerism of 9-Azabicyclo[4.2.1]nonan-l-ols
3.1 INTRODUCTION
The tropane alkaloid physoperuvine (170 171) was isolated from the roots of
physalis peruviana in 1976 (figure 3.1).® Early studies showed physoperuvine to be
isomeric with tropine (28) but the structure originally proposed, 3-(methylamino)-
cycloheptanone (172), was incorrect. The structure was revised in 1982 and was
assigned as 4-(methylamino)cycloheptanone (170 ^ 171) from spectroscopic
studies.^ This was later confirmed by X-ray crystallography of the hydrochloride
salt.^
MeN
HNMe HNMe
HO
(171) (172)Figure 3.1
The hydrochloride salt of physoperuvine existed in the bicyclic form (171.HC1)
only. Interestingly, from circular dichromism and IR spectroscopy, it was discovered
that an equilibrium existed in solution between the amino-ketone (170) and the
bicyclic hemiaminal (171) with an estimated ratio of 45:1. Interest in the
1-hydroxytropane skeleton has increased recently following the discovery of a new
class of 1-hydroxytropane derivatives in 1988.®* They were isolated from the roots of
calystegia sepium (convolvulaceae) and were recognised as growth enhancers of
nitrogen fixing bacteria. The structures were elucidated by Ducrot and Lallemand®^
two years later. The calystegines stimulated synthetic investigations which in turn
have led to a revival of interest in preparing physoperuvine itself. The synthesis of
these alkaloids and their derivatives is discussed fully in chapter 4. This chapter is
53
concerned with the higher homologues of physoperuvine, namely homotropan-1 -ols
(9-methyl-9-azabicyclo[4.2.1 ]nonan-1 -ols), norhomotropan-1 -ols (9-azabicyclo-
[4.2.1]nonan-l-ols) and derivatives thereof.
Smith^^^^ made preliminary investigations into the previously unstudied
higher homologues. The fiist target chosen was 4-aminocyclooctanone (176) (figure
3.2) which, if a tautomeric equilibrium existed, would in effect be the parent
norhomotropan-l-ol skeleton (177). The key intermediate for this synthesis,
4-hydroxycyclooctanone (173), was prepared via a singlet oxygen cycloaddition
reaction with cycloocta-1,3-diene (95). Successive tosylation to give (174),
substitution with azide ion, and hydrogenolysis of the azido-ketone (175) furnished
4-aminocyclooctanone (176).
HO TsO
(173)
Figure 3.2
TsCl/pyridine
65*
o (174)
HN
HO
(175)Hj^d/C
80*
NHo
(176)
Temperature-20°C-30°C-50°C
Ratio82 89 >95
1811<5
At room temperature the ^H and NMR spectra of (176) were broad,
suggesting that tautomeric interconversion was operating. At lower temperature the
tautomerism was slowed sufficiently on the NMR time scale to allow identification of
54
both the amino-ketone (176) and the bicyclic hemiaminal (177). Integration of
individual signals gave a quantitative measurement of the equilibrium ratio; this was
found to be temperature-dependent showing an increasing preference for the bicyclic
tautomer (177) with a decrease in temperature.
The N-benzyl derivative was produced from the previously described
amido-ketone (112). Protection of the ketone moiety using an acetal group, hydride
reduction and deprotection afforded 4-(benzylamino)cyclooctanone (178) in an
overall yield of 55% (figure 3.3). At -50°C, the NMR spectrum indicated that the
amino-ketone (178) was the major tautomer in equilibrium with 34% of the bicyclic
hemiaminal (179).
HNCOPh
Figure 33
HNCHgPh
3 steps
35%
CHzPhN
(178)HO
(179)
Temperature-50»C
Ratio 66 : 34
Attempts to prepare 7,8-dehydro derivatives of such hemiaminals starting from
unsaturated analogues of (173), via tosylation and substitution with azide, were
precluded because of the unexpected formation of a triazoline by an intramolecular
cycloaddition between azide and the double bond. Smith^^’ therefore used an
alternative approach to prepare unsaturated analogues starting from the amido-ketone
(180). Deprotonation with n-butyllithium generated an anion which was trapped with
2-methoxyethoxymethyl-chloride (MEM-Cl) giving the protected bicyclic ether
(181). This permitted hydride reduction of the benzoyl group (without simultaneous
reduction of the carbonyl group) to afford (182). Deprotection of (182) was achieved
using an excess of TFA yielding (184), which again existed in equilibrium with the
amino-ketone (183) as the marginally favoured component (figure 3.4).
55
HNCOPh COPh
"BuU/MEM-Cl
OMEM(180) (181)
CHzPhN
TFA.86*
HNCHgPh
Figure 3.4
CHzPhN
(183)Temperature
HO
Ratio 58 : 42
The difficulties encountered in removing N-benzyl groups from azabicyclic
sy stem s/^^ restricted the range of N-substituents that could be prepared using this
approach. In particular, methyl derivatives were unobtainable using an N-benzoyl
protecting group as an intermediate.
3.2 SYNTHESIS OF HOMOPHYSOPERUVINE,
NORHOMOPHYSOPERUVINE AND DEHYDRO-DERIVATIVES
A universal method of preparing N-methyl derivatives was sought which would
allow for the preparation of homologues of physoperuvine. The alkoxycarbonyl
group was chosen (rather than the amide) for the reasons discussed previously. The
a,p-unsaturated ketone (151) appeared to be an ideal substrate with which to employ
the MEM-Cl protection strategy (figure 3.5), as it was found to be in tautomeric
equilibrium with the bicyclic form (152). Treating (151 152) with base and
quenching the anion with MEM-Cl afforded the protected MEM-ether (185) in 73%
56
yield (figure 3.5). The and NMR spectra were complex because of the
presence of a pair of rotamers; the NMR spectrum showed the incorporation of a
MEM-ether with two signals at 6 3.35 and 8 3.38 corresponding to methoxy protons,
other signals were also consistent with the incorporation of the protecting group. The
NH signal was absent which suggested a bicyclic structure. The NMR spectrum
exhibited two diagnostic quaternary signals for the newly created bridgehead carbon
at 8 98.1 and 8 99.0 (two rotamers). Reduction of the carbamate (185) with lithium
aluminium hydride and deprotection of the crude amine with excess TFA afforded
N-methyl-9-azabicyclo[4.2.1]non-7-en-l-ol (187) in somewhat disappointing yield
(45%, two steps).
HNMeHNCCkCHnPh
O
73* (2) TFA 45*OMEM
COzCHzPhN
(185)
Figure 3.5
(152)
At -50°C
(187)
At room temperature both *H and NMR spectra were broadened indicating
the existence of a monocyclic / bicyclic equilibrium. The IR spectrum of (186 ^
187) included an absorption at 1705 cm"' characteristic of the a,P-unsaturated
carbonyl moiety of the monocyclic tautomer (186). The NMR spectrum showed
no evidence of either the a,P-unsaturated carbonyl carbon or bridgehead (COH)
quaternary carbon signals because of the coalescence of these widely separated
signals. However, measurement of the NMR spectrum at -50°C resulted in
57
sharpening of the spectrum and a quaternary bridgehead (COH) signal was now
visible at 6 94.8. Other signals were present which were consistent with the
bicyclic tautomer (187) as judged by comparison with the N-benzyl analogue (184).
No other signals were visible in the regions expected for signals unique to the
monocyclic tautomer (186) and the ratio was therefore estimated to be ^ : ^100 for
(186 187) respectively at -50°C. Unfortunately, the NMR spectrum at this
temperature was still broadened slightly and the signals due to the fractional
concentration of the monocyclic tautomer (186) were not resolved; a quantitative
estimate could not therefore be made. The observation of a carbonyl signal in the IR
spectrum at ambient temperature, coupled with the observation of only the bicyclic
tautomer (187) at reduced temperature, confirms a temperature-dependent
equilibrium.
Syntiiesis of the saturated form of (186 ^ 187) was of interest for two reasons.
Firstly this would be a first synthesis of the higher homologue of physoperuviire
(homophysoperuvine). Secondly, the establishment of a tautomeric equilibrium in
physoperuvine raised the question as to whether such a process would operate in
homophysoperuvine (190). It was decided to use the MEM-Cl protection method to
prepare this compound (figure 3.6). Thus, reaction of the previously prepared
HNMe
COgCHzPhHNCOnCHiPh
"BuLi/MEMO70*
OMEMO
(DÜAIH4(2)1TA
17*
(189)
Figure 3.6HO
(190)
58
a,p-unsaturated ketone (133) with base and MEM-Cl gave the bicyclic ether (188) in
70% yield. Both the and NMR spectra were very similar to those of the
unsaturated analogue (185), showing characteristic (COMEM) signals at 5 95.6 and S
96.3 (two rotamers). Reduction of (188) with lithium aluminium hydride and
acidification of the unpurified MEM-ether with an excess of TFA gave crude
homophysoperuvine (190), but in a yield of only 17%.
Only 21 mg of crude product was isolated using this method and it could not be
purified satisfactorily. Full spectroscopic characterisation and an investigation into
tautomerism involving the amino-ketone form (189) was not possible at this stage.
The major cause of the low yield was believed to be the high solubility of the amine
in the aqueous layer during extraction. Despite meticulous care, using the minimum
quantity of aqueous solution (saturated with sodium chloride) and repeated extraction
with ethyl acetate, the yield could not be improved.
The preparation of norhomotrop-7-en-l-ol (191) was thwarted because of the
same practical difficulties. Attempted acid or base hydrolysis of the carbamate (151
^ 152) resulted in highly polar products which could not be fully isolated or
identified. Application of the more modem deprotection reagent, iodotrimethyl-
silane,^ also failed giving only polymeric products (figure 3.7).
HNCOnCHzPh COgCHzPh
HO(152)
HO
Figures.?
Norhomotropan-l-ol (norhomophysoperuvine) (177) could be prepared
efficiently and simply from the saturated carbamate (133). Hydrogenolysis with
palladium on charcoal catalyst afforded (176 ^ 177) in near quantitative yield
(figure 3.8). The *H and NMR spectra were found to be identical to a sample
59
previously prepared by Smith.^^
HNCOzCHgPh ^ 2
H2mdÆ98*
HN
HO
(177)O o
(133) (176)Figure 3.8
3.3 AN EFFICIENT SYNTHESIS OF HOMOTROPAN-l-OLS
The use of a MEM-eÜier protecting group during the preparation of
homophysoperuvine (190) and homophysoperuv-7-ene (187) was only partially
successful as the yields were disappointing (17% and 45% respectively) and the
former was not characterised. A more efficient procedure was developed which
entailed direct oxidation of amino-alcohols, avoiding the necessity for a carbonyl
protecting group. It was hoped that oxidation of saturated (192) and unsaturated (193)
amino-alcohols, prepared from the respective carbamates (132) and (150) by hydride
reduction, would provide a more direct approach (figure 3.9). Both amino-alcohols
displayed identical spectroscopic properties to samples prepared by a different
route.® Corey’s® pyridinium dichromate (PDC) reagent was tried first as it was
reportedly a mild reagent which could be used in aprotic media. However, attempted
oxidation in both dichloromethane and DMF failed, resulting only in the isolation of
starting material. A similar outcome resulted when the closely related pyridinium
chlorochromate^^ (PCC) was employed. Barium manganate,*^ manganese dioxide^
and tetrapropylammonium perruthenate^ (TPAP) all failed similarly. Despite these
early failures, the use of Jones reagent in a procedure which did not involve an
aqueous work up gave good yields of both homophysoperuvine (189 ^ 190) and
homophysoperuv-7-ene (186 ^ 187).
60
HNCOzCHzPh HNMe
UAIH492*
Jonesreagent
69*
OH(132)
OH(192)
MeHNMe
HO(190)(189)
HNCOoCHfh
OH OH
HNMe
100 at -30=C
Me
Jonesreagent
63*
HOo(186)(150) (193)
Figure 3.9
The successful oxidation of (192) was confirmed by the isolation of a product
having identical spectroscopic properties to a sample prepared using the MEM-Cl
protection method. The ^H and NMR spectra of (189 «== 190) bore a close
resemblance to (186 ^ 187), being broad at room temperature with no quaternary
carbon signals visible in the NMR spectrum. The IR spectrum displayed an
absorption at 1695 cm \ indicative of a non-conjugated carbonyl group, hence the
monocyclic tautomer (189) was present at room temperature. Cooling the sample to
-30°C made little difference to the *H NMR spectrum, but the ^ C NMR spectrum
was now much sharper with the appearance of a new quaternary carbon signal at 5
92.0 together with other signals characteristic of the bicyclic hemiaminal (190). The
carbon shifts of both the quaternary bridgehead (6 92.0) and the tertiary bridgehead (5
58.1) in the ^ C NMR spectrum were in close accord with published data for
calystegine which had values of S 93.0 and 5 54.0. Analysis of the NMR spectra
led to a similar conclusion to that reached for the unsaturated derivative (186 ^ 187),
namely that a temperature-dependent process was operating and that the equilibrium
was weighted very heavily towards the bicyclic tautomer (effectively ~0 : ~100 for
61
(189 ^ 190) respectively) at -30°C. No other signals that might correspond to the
monocyclic tautomer were observed. Discussion of the relative tautomer preferences
will follow a description of similar studies of closely related epoxy-derivatives.
3.4 SYNTHESIS OFEXO-7,8-EPOXYHOMOTROPAN-l-OLS
In order to prepare novel higher homologues of calystegines and extend the
study of tautomerism processes in derivatives of homophysoperuvine, the introduction
of oxygen functionality in the 7,8-bridge was deemed to be important. One possible
approach would be elaboration of the etheno-group of 7,8-dehydrohomophyso-
peruvine (187), but because practical difficulties were encountered in the handling of
such amines, this was not considered to be a feasible approach. A more suitable
substrate was sought; with the high-yielding and stereoselective epoxidation reaction
of the previous chapter in mind, it was decided to attempt an epoxidation of the
N-protected 4-aminocyclooct-2-enone (151) which had been found previously to be in
equilibrium with the bicyclic tautomer (152). It was anticipated that epoxidation
would occur via the more labile double bond of (152) rather than that of the less
reactive a,P-unsaturated ketone. This was indeed the case as epoxidation of (151 ^
152) with MCPBA furnished the exo-epoxide (194) as the only product in good yield
(figure 3.10). If epoxidation of (151) had occurred then a mixture of stereoisomers
may have resulted.
HNCOgCH^Ph COgCHzPh COzCHzPhN N
HO(152)
HO
Figure 3.10
The ^H and NMR spectra were consistent with the bicyclic tautomer (194)
only and, as a consequence of slow rotation about the N-CO bond, two sets of signals
62
were observed. Signals due to both rotamers were well resolved at room temperature
and there was no evidence of any monocyclic epoxy-ketone tautomer. In the
NMR spectrum, the epoxide protons of the major rotamer appeared as two doublets at
8 3.33 and 8 3.48 which had a mutual vicinal coupling of 3.2 Hz. Similar signals for
the minor rotamer were also present. No vicinal coupling between the Hg-bridgehead
and the H^-epoxide protons was seen. This is in close agreement with the analogous
1-methyl epoxide (165), which displayed only a small coupling of 0.4 Hz for J5 7.
The NMR spectrum of (195) included a characteristic quaternary (COH) signal at
891.4.
Having prepared the epoxide (194), it was expected that reaction with hydride
would simultaneously reduce the carbamate to an N-methyl group and open the
epoxide to yield a mixture of regioisomeric alcohols (195a,b). This was not the case:
treatment of (194) with lithium aluminium hydride in THF at reflux resulted in
reduction of the carbamate only, giving the novel epoxy-amine (196) in good yield
(figure 3.11).
MeN
MeN
COgCHzPhN
HO(194)
HOHO
HO HO
MeN
HOFigure 3.11
This was confirmed by the *H NMR spectrum which displayed very similar
epoxide signals to (194), with two doublets at 8 3.34 and 8 3.38 exhibiting the
characteristic vicinal coupling of 3.3 Hz. The formation of an N-methyl group was
63
shown by a singlet at S 2.52 integrating to three protons in the NMR spectrum and
a quartet at 6 28.9 in the NMR spectrum.
Similar chemoselectivity was observed when (194) was hydrogenolysed to
(197) using a palladium on charcoal catalyst (figure 3.12). The epoxide protons (H^ g)
HN N
HO(194)
HO(197)
Figure 3.12
were magnetically equivalent in CDCI3 solvent, but in Dg-acetone these were
resolved and the *H NMR spectrum displayed two doublets at 5 3.27 and 6 3.31 (J7 g =
2.7 Hz). The NMR spectrum showed a bridgehead (COH) signal at 5 91.2 in
close analogy to (196).
There was no spectroscopic evidence that either (196) or (197) existed in
tautomeric equilibria at room temperature, in agreement with observations on (194).
The crude *H and NMR spectra of neither (196) nor (197) showed any indication
at all of epoxide ring opening. This stability is suprising in view of the general
lability of epoxides, and is not fully understood. An investigation into the relative
reactivities of exo- and endo- epoxides is made later (chapter 6).
3 J CONCLUSION
The ratios of monocyclic.bicyclic tautomers in the saturated, unsaturated and
epoxyhomotropan-l-ol derivatives are summarised in table 1. The N-protected
epoxide (256) having a frons-stereochemistry between the nitrogen and the epoxide,
prepared later (chapter 5.3), is included for comparison.
The tautomeric ratios of primary and secondary amino-ketones in the saturated
64
HX
Table 1. Ratios of monocyclic (M/C) : bicyclic (B/C) tautomers
X HX
OH Û OH
X Compound Temp M/C B/C 13c NMR(5)("Q M/C B/C
Saturated series C4 CO Cl C(NCOzCHiPh (133^ 134) 25 71 29 50.8 217.0 92.9 55.4NMe (189 — 190) -30 ~0 -100 - - 92.0 58.1NCHgPh (178 ^ 179) -30 70 30
-40 68 32-50 66 34 56.1 218.6 92.2 54.2
NH (176^ 177) -20 18 82-30 11 89 50.7 218.8 93.1 52.4-50 5 95
Unsaturated seriesNCOgCHzPh (151^ 152) 25 37 63 50.6 203.4 95.7 60.8NCH^Ph (183 ^ 184) -55 58 42 55.3 203.3 94.7 60.1NMe (186 ^ 187) -50 -0 -100 " - 94.8 62.9Epoxide seriescw-NCOgCHzPh (194) 25 0 100 - - 91.4 58.0trans-NC02CH2Ph (256) 25 100 0 52.3 206.2 - -
NMe (196) 25 0 100 - - 89.6 55.3NH (197) 25 0 100 - - 91.2 54.6
and unsaturated series could not be measured at ambient temperature (witii the
exception of the epoxides) because of rapid interconversion between tautomers and
the values in all these cases refer to temperatures of -20°C or below. The observation
of carbonyl absorptions in the IR spectrum at room temperature, particularly for (189
^ 190) and (186 ^ 187) (which appear to be totally bicyclic at low temperature)
indicates that a temperature-dependent process is operating. The observation of
variations of ratios with temperature means that comparisons between values
65
measured at different temperatures should be treated with caution. Interconversion of
carbamate-protected amines was considerably slower (possibly because of the
reduced nucleophilicity of the nitrogen) and as a result these ratios could be measured
at room temperature.
The heavy preference for the bicycle (190) at -30°C corresponds closely with
the 98:2 ratio measured for physoperuvine itself, both at ambient temperature^®^ and
at reduced temperature®^ (chapter 4). The similar preference for the bicyclic tautomer
measured for the nor-system (176 ^ 177) is also in agreement with the ratios
measured in the lower homologues i.e norphysoperuvine®^ and calystegines.®^ The
N-benzyl derivative differed somewhat showing a preference for the monocyclic
tautomer (178) but, in accord with (176 ^ 177), a decrease in temperature increased
the proportion of bicycle (179).
The introduction of a C7 g-double bond into the N-benzyl derivative to give the
dehydro-analogue (183 ^ 184) resulted in a minor perturbation towards the bicyclic
tautomer (185). This effect was exaggerated in the analogous carbamates with
saturation favouring the monocyclic tautomer (133) and unsaturation favouring the
bicyclic tautomer (152) at room temperature. In contrast, N-methyl derivatives were
both effectively bicyclic, with or without a tt-bond. The inclusion of an exo-epoxide
group into the C7,Cg bridge has a marked effect on the position of equilibrium with
the carbamate, N-methyl and nor- derivatives existing as the bicyclic tautomer only.
The stereochemistry of this epoxide has a marked effect on tautomer preferences, with
the tran^-epoxide (256) existing in the anomolous monoeyelic form only.
It is evident from these ratios that the position of equilibrium is dependent on
both the nitrogen substituent and the substituents in the 2-carbon bridge. The
incorporation of a double bond would be expected to stabilise the monocyclic
tautomer considerably because of resonance in the a,P-unsaturated ketone group. As
a consequence it might be anticipated that an equilibrium shift towards the
a,p-unsaturated monocyclic tautomers would be observed. In fact there is only a
marginal shift for the N-benzyl derivative (183 ^ 184) and the shift is actually
66
reversed for the carbamate (133 ^ 134) with the saturation of the double bond
decreasing the proportion of monocyele (151). A double bond therefore appears to
induce some stabilisation into the bicyclic tautomer. There is strong evidence®® that
rigid bicyclic systems such as 7-azabicyclo[2.2.1]hept-2-ene and hepta-2,5-diene
derivatives experience a stabilising a-n* interaction between the bridging C-N bonds
and the rc-system, and such an effect would conveniently rationalise the observed
change in ratio. However, this effect seems to be absent in larger, more flexible ring
systems; NMR studies®® have shown that the interaction is heavily attenuated in
trop-6-enes and plays no detectable part in the homotrop-7-ene ring system. The
influence of temperature on the position of equilibrium is a further factor. The
increase in the proportion of monocyclic tautomer with an increase in temperature is
to be expected in view of the increased entropy which results from the formation of
the monocyclic over the bicyclic tautomer. The overall balance of factors influencing
the relative stability of the two tautomers is clearly complex and cannot easily be
discerned. A similar somewhat irrational pattern is also observed in the lower
homologues, tropan-l-ols, discussed in the next chapter.
67
CHAPTER 4
8-Azabicyclo[3.2.1]octan-l-ols; Synthesis of Physoperuvine, Norphysoperuvine & Dehydro-derivatives
4.1 INTRODUCTION
Of the numerous alkaloids based on the tropane and nortropane skeleta, only
physoperuvine (171) and the calystegines (198 - 200) have a hydroxy group at the Cj
position, thereby having the potential to exist in an equilibrium between monocyclic
amino-ketone and bicyclic hemiaminal forms. The three calystegines^ Bj, 82 and A3
(figure 4.1) which have been isolated recently are dissimilar to physoperuvine as they
exist as the bicyclic tautomers only.^ No explanation for this observation was
offered by Lallemand,®^ other than the presence of substituents favouring a shift
towards the bicyclic tautomer.
H H HN N N
HO
HOHO OHHOHOHO HO
OHOH OHBi (198) 82 (199) A3 (200)
Figure 4.1
Physoperuvine has been prepared only twice, with the pivotal step in both cases
based on diazomethane ring expansion of 4-amino cyclohexanone derivatives. As
part of a study concerned with the structural assignment of physoperuvine. Finder®®
reported the first synthesis in 1982 (figure 4.2). 4-(Methylamino)phenol sulphate
(201) was catalytically reduced, using the procedure of Heckel and Adams,^°° to the
cyclohexanol (202) as a mixture of cis and trans isomers. Oxidation to the
amino-ketone (203) and subsequent homologation with diazomethane afforded
physoperuvine (170 ^ 171). Although the final step was high yielding, earlier steps
were much less efficient, resulting in an overall yield of less than 10%.
The second synthesis was published by Lallemand*® in 1992 and was based on
68
HNMe HNMe HNMe
OH ÏH2SO4
Figure 4.2 HNMe
Physoperuvine ®
(171) (170)
a similar ring expansion of a cyclohexanone derivative as outlined in figure 4.3.
4-Aminocyclohexanol hydrochloride (204) was sequentially benzylated to (;^5) and
then oxidised to the amino-ketone (206). Treatment with diazomethane afforded
NH2.HCI HNCHgPh HNCHoPh
PhCHCVNa/A
64%
Jones reagent
50%
OH (204) OH (205) O (206)
y CH2N2.29%
NH
HO(209)(210)
HNCHoPh
HgPd/C
Norphysoperuvine
Figure 4.3
(207)
Physoperuvine
(208)
69
4-(benzylamino)cycloheptanone (207). Studies were not reported concerning the
possible tautomerism of (207) with the hemiaminal (208). No procedure was detailed
for the conversion of (207) to physoperuvine, but the overall efficiency could not
exceed 10% because of the low yielding oxidation and homologation steps.
Alternatively, hydrogenolysis of (207) afforded norphysoperuvine (209 ^ 210)
which was reported to exist as a reversible polymeric stmcture.
Other investigations into tropan-l-ol and nortropan-l-ol systems have
concentrated on the preparation of calystegines. Preliminary investigations were
made by Lallemand,^^ who devised an approach to the nortropan-l-ol system, but
considerable advancements have been made since. The synthesis^® of racemic
calystegine A3 is discussed in detail here (figure 4.4). The benzyloxycarbonyl-
protected amino-ketone (211), prepared using a similar procedure to that outlined in
figure 4.3, was the key starting material but an alternative ring enlargement strategy
was employed which made accessible a- and p-hydroxylated compounds later in the
synthesis. Cyclopropanation of the derived silylenol ether (212) and ring opening of
the cyclopropane (213) furnished the a,p-unsaturated ketone (214) in an overall yield
of 63% from (211). Epoxidation using alkaline hydrogen peroxide gave a 1:1 mixture
of epoxides, (215) and (216), which were separated by HPLC. Two tran^-diols, (217)
and (218), were then obtained by acid-induced epoxide ring cleavage and the former
was hydrogenolysed to give natural calystegine A3 (200) Although this is only a
moderate yielding synthesis (because of the formation of two pairs of isomers) it
provides a means of preparing the non-natural trans-diaxial analogue (220).
Interestingly, whereas calystegine A3 existed as the hemiaminal tautomer only, NMR
investigations showed the non-natural analogue (219) was in equilibrium with the
monocyclic tautomer (220). The position of this equilibrium was not reported, but the
observation points to the fine balance that exists between tautomers.
70
HNR HNR HNR
MegSiOŒtgN CHglg.Zn/Ag
o(211)
MegSiO(212)
MegSiO(213) (1)Peag
(2)NaOAcŸ 59% overall
HNR HNR HNR
(215) (214)
R = CO,CH,PhHNR HNR
OH ■■■ OH
OH OH
(218)Figure 4.4
(217)
Calystegine Ag
(200)NH
OH
HO HOHO
OHOH
HO (219) (220)
Further syntheses of calystegines include an elegant method by Lallemand^®^ of
converting D-glucose into (-)-calystegine B2 based on the previously described
cyclopropanation to homologate a cyclohexanone derivative. Other homochiral
syntheses^® of both natural and non-natural calystegine B2 have been undertaken,
using isoxazoline chemistry, again starting from D-glucose. To date, however, a high
yielding preparation of 8-azabicyclooctan-l-ols has not been reported, neither have
detailed variable-temperature NMR studies been made of tautomeric equilibria of
these systems. This chapter describes an effective synthesis of physoperuvine and
simple derivatives thereof. An investigation into the tautomeric preferences is also
made.
4.2 SYNTHESIS OF PHYSOPERUVINE AND NORPHYSOPERUVINE
The successful synthesis of homophysoperuvine (191) from cycloocta-1,3-diene
(95) provided the impetus for a parallel synthesis of physoperuvine from
cycloheptadiene as outlined in figure 4.5. The key cycloadduct (221) was thus made
from (43) and benzylnitroso formate, with the use of a rigorously purified sample of
the nitroso precursor, benzyl-N-hydroxycarbamate, resulting in a near quantitative
yield. The spectroscopic properties of (221) were identical to those of a sample
prepared by a previous route.^^ Reduction with diimide to (222), followed by
reductive cleavage of the N-O bond afforded (223) in 98% yield; subsequent
reduction with lithium aluminium hydride gave the amino-alcohol (224). The ^H and
NMR spectia of these compounds bore a close resemblance to those of the higher
homologues. The final step used the Jones procedure (as described previously for
homophysoperuvine) which did not use an aqueous work-up and provided
physoperuvine (170 ^ 171) in an overall yield yield of 79% from
cyclohepta-1,3-diene.
The oxidation of (224) was confirmed by a reduction in relative molecular mass
of 2 as shown by mass spectroscopy and the presence of a weak absorption at 1695
72
(43)
RczCOgCHzPh
HNMe
97*
(221)(222)
Na/Hg98*
HNMe HNR
Jones reagent
94*
UAIH494*
(224) HO (223)
98
HO(171)
Figure 4.5
cm * in the IR spectrum, characteristic of a non-conjugated carbonyl stretching
frequency. As with previous Jones oxidations the amine moiety was undamaged in
this reaction. The specific observation of individual tautomers was not possible using
NMR spectroscopy at ambient temperature because of relatively rapid
interconversion; neither the *H nor the NMR spectra were fully analysable being
broad and complex. However, the preference for the bicyclic tautomer (171) became
apparent when the NMR spectrum was recorded at low temperature (figure 4.6).
At -50°C, a characteristic quaternary signal appeared at Ô 88.8 and was assigned to the
(COH) bridgehead. After extended spectral accumulation (6000 scans) minor signals
confirmed the presence of a small proportion of the monocyclic tautomer (170) but
73
the carbonyl signal was too weak (or broad) to be observed. A ratio of 2:98 was
estimated from integration of the signals, in good agreement with the earlier
estimate.^®
Norphysoperuvine (209 ^ 210) was prepared by hydrogenolysis of (225)
which was prepared, in turn, by Jones oxidation of (223) as outlined in figure 4.6.
The N-protected derivative (225) appeared to be totally monocyclic at ambient
temperature as shown by a carbonyl signal at 8 213.7 assigned to the Cj carbonyl of
the monocyclic tautomer and the absence of a charateristic (COH) quaternary carbon
signal in the 8 85 - 90 region of the * C NMR spectrum. The *H NMR spectrum was
also consistent with only (225) showing no evidence of the slow N-CO rotation
typical of the bicyclic carbamates. Hydrogenolysis of (225) gave norphysoperuvine
(209 ^ 210) in 92% yield (overall yield 79% from cycloheptadiene). The *H and
HNR
HO
HNR
Jones reagent _
96*
(223) (225)100 : - 0 (25«C) (226)
R = C02CH2Ph
Figure 4.6
NH
Hg/PWC92*
O (209)0 :-100 -(50^)
(210)
Norphysoperuvine
* C NMR spectra of (210) were very similar to those observed for
homophysoperuvine and physoperuvine being broad and unresolved at 25°C. On
cooling to -50°C only one tautomer was observed, with a new signal appearing in the
74
*^C NMR spectrum at 8 89.5 assigned to the bicyclic tautomer (210). Some
precipitation occurred from solution at this temperature so extended scanning to
increase the signal to noise ratio was not an option. However, all spectroscopic data
pointed to the presence of only one tautomer and the ratio of (209 ^ 210) was
therefore estimated to be ~0 : ~100 respectively. Neither *H nor NMR spectra
suggested any evidence of the polymerisation reported previously.*®*
4.3 DEHYDRO-DERIVATIVES OF PHYSOPERUVINE
A decision was reached to apply the oxidation of allylic alcohols to prepare
dehydro-derivatives of physoperuvine. This would provide a comprehensive range of
tropan-l-ol derivatives which could be compared with the homologous
homotropan-l-ols and may offer some indication of the effect of cycloheptane
compared with cyclooctane ring size on tautomer preferences. The benzyloxy-
carbonyl-substituted compound (229 ^ 230) was synthesised by cleaving the N-O
bond of the cycloadduct (221), with subsequent mild oxidation of the allylic alcohol
(227) using barium manganate (figure 4.7). The N-methyl analogue (231 ^ 232) i.e
6,7-dehydrophysoperavine was prepared from the corresponding allylic alcohol (228),
a product of hydride reduction of (227). Oxidation of (228) proved more difficult
than anticipated; the use of an identical Jones procedure to that described for the
higher homologue resulted in the isolation of starting material with a minor quantity
of polymeric material, which was probably formed during a final extended Soxhlet
extraction. Various other o x i d a n t s ^ w e r e tried but all failed to achieve the
desired oxidation. It was postulated that a possible failure of the Jones oxidation was
due to the protonation of the amine moiety of (228), generating an insoluble salt
which precipitated from solution and was thus unreactive. Successful oxidation was
finally achieved by modifying the reaction conditions slightly; the use of a more
dilute solution acidified with TFA (to produce a soluble salt) prior to the addition of
chromic acid afforded (231 ^ 232) in 40% yield after aqueous work-up.
75
Unfortunately the product was unstable, decomposing during chromatography, and as
a result a pure sample was not obtained.
HNR
Na/Hg
90*
HNMe
UAIH493*
OH(221) (227)
OH (228)
Jones/TFA, 40*
HNMeBaMnO^
79*
HNR
(231) HO^ Major ; Minor (232)
R at25«C
100 : ~ 0
R = COzCHjPh
Figure 4.7
(229) (230)
The NMR spectra of the carbamate derivative (229 ^ 230) were consistent
with its existence solely as the the monocyclic tautomer (229) at 25°C, and the
NMR spectrum showed a characteristic carbonyl signal at 8 202.9. The olefmic
signals in the *H NMR spectrum at 8 5.92 and 8 6.34 displayed a mutual vicinal
coupling of 12.4 Hz which, after comparison with the *H NMR spectrum of the
homologue (151), also confirmed that the monocyclic tautomer was present.
Attempts to deprotect (229) to give the nor-deiivative using acid hydrolysis were
unsuccessful.
Detailed NMR studies on 6,7-dehydrophysoperuvine (231 232) could not be
performed because of its instability and limited availability. However, despite the
76
presence of minor impurities, the NMR spectrum was very similar to that of the
analogous carbamate (229) showing a vicinal coupling between olefmic protons of
10.2 Hz. The IR spectrum showed a diagnostic absorption at 1665 cm** due to the
a,p-unsaturated carbonyl and it was thus concluded that the major component at 25°C
was the monocyclic tautomer (231). Insufficient sample was available for a
satisfactory NMR spectrum to be obtained, so a more quantitative estimate could
not be made.
The instability of the N-methyl amine restricted the spectroscopic investigations
into tautomer preferences considerably. It was decided to synthesise a different
N-substituted analogue in the hope of increased stability which might permit a more
detailed NMR study. Howarth^^ had previously prepared cis-4-(benzylamino)cyclo-
hept-2-enol (56) and a sample of this was oxidised to (233 234), again utilising an
initial protonation of the amine moiety with TFA (figure 4.8). Solutions of (233 ^
234) did decompose slowly on standing, but the yield of the reaction was better than
that obtained for the N-methyl analogue (presumably because of increased solubility
in the organic phase during work-up) and as a consequence the amine could be
studied by NMR spectroscopy.
PbHgC CHgPhN N
Jones, TFA
oOHHO HO
(56) (233) Equatorial (234a) Axial (234b)Figure 4.8
The *H NMR spectrum of the benzyl analogue was similar to (231) at 25°C and
the signals were consistent with the presence of the monocyclic tautomer (233). The
olefinic protons were observed at 8 5.97 and 8 6.56 with a mutual vicinal coupling of
12.3 Hz. Double irradiation experiments confirmed the relationship between the
77
protons on C2, C3 and C4 and a four-bond coupling was observed through the
carbonyl group between protons on C2 and C7 (J = 1.0 Hz). At ambient temperature,
the spectrum indicated that the monocyclic tautomer was the major component, but
on closer inspection additional signals could also be discerned; as the sample was
known to contain minor impurities it was not possible to assign these to the bicyclic
tautomer. However, further signals appeared as the temperature was decreased to
-50°C, including a broad carbonyl signal at 8 204.8 assigned to the monocycle (233).
This behaviour indicated the existence of a tautomeric equilibrium and, although the
spectra could not be fully interpreted, strongly suggested a heavy preference for the
monocyclic tautomer. The difficulties associated with interpreting the
low-temperature NMR spectra of the benzyl derivative may well be due to a second
temperature-dependent process, namely inversion at nitrogen. This has already been
shown^^ to cause spectral broadening for N-benzyl nortropane and N-benzyl
nortrop-6-ene over a similar temperature range, even though both of these compounds
show a heavy preference for the equatorial invertomer (234a) rather than the axial
invertomer (234b).
4.4 CONCLUSION
The ratios of monocycliczbicyclic tautomers observed for both the saturated and
unsaturated derivatives are summarised in table 1. Two epoxy-ketones prepared later
(chapter 6) in connection with other work, along with 4-benzylaminocycloheptanone
(207 ^ 208) synthesised by Howarth^^ are included for comparison.
Both the saturated (225 ^ 226) and unsaturated (229 ^ 230) carbamates
existed as the monocyclic tautomers only. This was in contrast to the situation for the
homologous homotropan-l-ols, where a mixture of tautomers was present. Also, both
cis- and rran^-epoxy-carbamates (269) and (276) were monocyclic only, whereas the
higher homologous showed a variation from being completely bicyclic to completely
monocyclic depending on the stereochemistry of the epoxide. It can be concluded
78
Table 1. Ratios of Monocyclic (M/C) ; Bicyclic (B/C) Tautomers
HX HX
HOO
HO
Compound Temp M/C : B/C
C O
:% NM R(5) M/C B/C
S a tu ra te series C4 CO Cl C6NCOgCHzPh (225^226) 25 -100 -0 52.7 213.7 - -NMe (170—171) -50 2 98 - - 88.8 58.1NCH^Ph (207^208) -50 minor major - - 88.9 56.2NH (209^210) -50 -0 -100 - - 89.5 53.3Unsaturated seriesNCOgCHzPh (229^230) 25 -100 -0 51.6 202.9 “ -NMe (231^232) 25 major minor - - - -NCHgPh (233^234) -50 major minor 56.7 204.8 - -Epoxide seriescw-NCOgCHzPh (269) 25 100 0 51.2 208.5 - -frmw-NC02CH2Ph(276) 25 100 0 48.8 209.3 - -
that, for the saturated and unsaturated derivatives at least, an important factor
governing tautomeric preferences is ring size i.e. the instability of the tropan-l-ol
structure relative to homotropan-l-ol.
Both physoperuvine (170 ^ 171) and norphysoperuvine (209 ^ 210) bore a
close resemblance to their respective higher homologues, showing a marked
preference for the bicyclic forms. However, the incorporation of a double bond into
physoperuvine to give 6,7-dehydrophysoperuvine (231 ^ 232) resulted in a shift
towards the monocyclic tautomer. This is surprising as the reverse shift was
previously seen for the respective higher homologues; it would appear that the
apparent decreased stability upon incorporation of a shorter double bond into the
already strained tropan-l-ol structure is the overriding effect here. A similar pattern
79
was observed for the N-benzyl derivatives (207 — 208) and (233 — 234).
Despite the difficulties encountered in measuring the tautomeric ratios of these
compounds, some patterns can be discerned between derivatives. The factors
controlling preferences are complex and not fully understood. Nonetheless, the
syntheses of physoperuvine and norphysoperuvine were efficient. Further work on
dehydro-derivatives was of limited value because of their instability.
CHAPTER 5
Stereoselective Oxygenation of Homotropane
5.1 INTRODUCTION
The synthesis of homotropane, norhomotropane and some C}-substituted
derivatives has been described, with the previous difficulties of removing
N-protecting groups having been overcome by the use of N-alkoxycarbonyl groups.
Oxygenation at the C2-position of homotropane was achieved by Cope,^* who
described the ring expansion of tropinone, and by Hobson '* who solvolysed
2-chlorohomotropanes. Other derivatives are rare and C .Cg oxygenation is unknown.
Methods of preparing unsaturated derivatives are known,^* but these have not been
extended to the synthesis of oxygenated derivatives. The scarcity of these compounds
is suprising in view of the potent activity of anatoxin-a and the well established
physiological properties of oxygenated tropanes.^-^ With this in mind an investigation
of the functionalisation of the 2-carbon bridge was considered a priority. The
stereoselective introduction of both hydroxy and epoxide groups is of key importance
in approæhes to a range of novel homologues that may possess similar physiological
properties to tropane alkaloids.
5.2 SYNTHESIS OF EJrO-7,8-EPOXYHOMOTROPANE
So far, a strategy to systems bearing bridgehead substituents has been
developed; both 1-methyl (165) and 1-hydroxy (194) derivatives of the N-protected
7,8-epoxyhomotropane ring system have been synthesised from the respective
1-substituted alkenes, with exclusive exo-stereoselectivity in each case (figure 5.1).
COgCHzPh COzCHzPhN N
n R = Me (165), 94%R = OH (194), 79%R = H (235)
Figure 5.1R
81
An obvious approach to the exo-epoxyhomotropane (235) was therefore epoxidation
of the homotrop-7-ene without a bridgehead substituent. This strategy was also
attractive since it allowed the straightforward deprotection of (194) at the conclusion
of the synthesis without reduction of the epoxide group.
A modification of the previously described intramolecular displacement
approach to homotropane^* was employed to prepare the alkene (238). Although
secondary amines have proved superior to carbamate groups in such cyclisations, an
N-protected derivative was necessary to facilitate a successful epoxidation reaction, in
view of the difficulties encountered by Fodor*^ whose reported epoxidation of an
unsaturated amine does not proceed efficiently. Figure 5.2 outlines the approach to
(238) starting from the known intermediate (150).
HNR HNR HNR
"BuLi/TsQ #
66%
LiCl/DMSO 1
45%
OH OTs
(150)
R = C02CH2ph
Figure 5.2
(236) (237)NaH
(239)
RN
(238)
THF/RefluxTHF/DME/60^
10%3%
28%40%
It was hoped that tosylation of the 4-hydroxy group and subsequent
displacement with chloride ion would provide the required trans-1,4 stereoelectronic
arrangement for cyclisation. Difficulties were encountered with these
transformations, firstly because the tosylate (236) was unstable and could only be
prepared in modest yield. Secondly, chlorination of (236) using lithium chloride in
82
DMSO solvent^® resulted in épimérisation at the 4-position with the chloride (237)
being isolated as an inseparable mixture of stereoisomers in a 7:3 trans:cis ratio
(calculated from NMR signal integrations), again in a somewhat disappointing
yield. Attempts to invert the hydioxy group of (ISO) to a chloride by other methods
(such as Smith’s * use of thionyl chloride or triphenylphosphine in carbon
tetrachloride^®^) failed giving complex reaction mixtures. Nevertheless, base-induced
cyclisation of (237) was achieved by refluxing with sodium hydride in THF solvent
giving the homotrop-7-ene (238) in 28% yield together with a smaller quantity of
aziridine by-product (239), formed by a competitive 1,2-cyclisation. Examination of
the NMR spectrum of the crude reaction product showed that benzyl alcohol was
also formed as a result of deprotection of the benzyloxycarbonyl moiety. Milder
conditions were therefore used: heating (237) at 60°C (rather than refluxing) in
THF/DME solvent increased the yield of (238) to 40% and also suppressed the
formation of aziridine. The NMR spectra of (238) were simplified somewhat because
of the associated symmetry but, typical of bridged bicyclic carbamates, displayed
duplication of signals as a result of slow N-CO rotation. The NMR spectrum
showed two signals at Ô 5.72 and 8 5.75 for the olefmic protons with mutual coupling
of 6.8 Hz. Each proton showed further coupling (J = 2.3 Hz) to the bridgehead
protons. These values were similar to those observed for the Cj-methyl analogue
(165) prepared previously (chapter 2.5). The ^H NMR spectrum of (239) displayed
chaiacteristic aziridine protons at 8 2.63 and 8 3.07 with a mutual coupling of 6.3 Hz.
The NMR spectrum showed two (CH) aziridine signals at 8 40.0 and 8 44.4 which
fits well with values of 8 42.4 and 8 47.6 for the analogous aziridine (58), a by-product
of a similar cyclisation in the tropane series.^^
Epoxidation of (238) using MCPBA afforded the anticipated exo-epoxide (235)
in high yield (figure 5.3). As a result of slow N-CO rotation, the ^H NMR spectrum
of (235) included two doublets at 8 3.36 (J = 3.0 Hz) and 8 3.37 (J = 3.0 Hz) for the
epoxide protons and two further doublets at 8 4.35 (J = 6.2 Hz) and 8 4.41 (J = 6.2 Hz)
for the bridgehead protons. The observation that the epoxide and the bridgehead
83
protons were not vicinally coupled fits well with observations for the Ci-methyl (165)
and Cl-hydroxy (194) analogues and confirms the stereochemistry of the epoxide.
COzCHzPhN
(238)
MCPBA/CH2CI2
Figure 5.3
COgCHzPhN
The investigation of facial selectivity using reagents other than MCPBA was of
interest. Earlier studies have shown that hydroboration-oxidation of norbomene
afforded the exo-alcohol as the sole product*® and, with this in mind, it was decided
to use borane chemistry as a means of introducing an exo-hydroxy group into the
homotropane skeleton. Using the procedure of Wilt and Narutis,^® (238) was treated
with 0.6 equivalents of a borane:THF complex which resulted in a good yield of the
exo-alcohol (239) as the only product. Hydrogenolysis with palladium on charcoal
catalyst afforded norhomotropan-T^-ol (240) in near quantitative yield (figure 5.4).
COzCHgPhN
COzCHgPhN
BH3:THF/H202 HO H2/Pd/C HO96%
(238) (240) (241)Figure 5.4
The incorporation of H2O during the hydroboration-oxidation step was
confirmed by an increment in relative molecular mass of 18. The *H NMR spectrum
of (240) displayed the absence of olefinic signals and the appearance of a broad
exchangeable signal at 8 2.89. Both the *H and * C NMR spectra were complex
because of restricted rotation about the N-CO bond. However, these spectra were
greatly simplified after hydrogenolysis; the ^H NMR spectrum of (241) now showed a
84
doublet of doublets at 5 4.12 (J = 6.2, 2.8 Hz) corresponding to the a-hydroxy proton
H7. These two couplings were to protons on Cg and no vicinal coupling between the
H7 proton and the Hg-bridgehead proton was observed. The absence of measurable
coupling between H7 and Hg was also noted for the H7 and Hg protons of the
exo-epoxide (235) which strengthens the assignment of P-hydroxy stereochemistry.
The structure of (241) was further endorsed after comparison of NMR coupling
constants with those measured for ring-opened exo- and endo-epoxyhomotropanes
described later (section 4).
Although a strategy for the synthesis of N-protected exo-epoxyhomotropane
(235) and norhomotropan-7 P-ol (241) has been described, the efficiency was low
owing principally to the presence of a double bond during the inversion of (236) and
cyclisation of (237). A more attractive approach would therefore be to introduce the
epoxide group at an earlier stage of the synthesis which would prevent the formation
of rearrangement and decomposition products, and should significantly improve the
overall yield. Such a method was discovered fortuitously in an attempt to open the
epoxide (194) to the alcohol (242), as outlined in figure 5.5. It has been shown
RN
HO
OH
(242)
O
RN
MeN
OH
1 (194)
UAIH471$
o
OH
(196)
R = C02CH2Ph
Figure 5.5
HNR
(243)
HNRNaBH4, A
98$
Jones reagent
95$
(244)
85
previously that the epoxide (194) was remarkably stable and could be reduced with
lithium aluminium hydride to (196), with neither (194) nor (196) showing any
evidence of the presence of their respective monocyclic epoxy-ketones at room
temperature. Various methods were employed in attempts to ring open the epoxide
(194); simple acid or base hydrolysis failed, giving mixtures of starting materials and
deprotected products only. Numerous reductive ring opening methods have been
reported in the literature.Brown^®^ used lithium metal in ethylenediamine to
reduce norbomene oxide to norbomanols whilst Yankar^^® reduced epoxides with a
combination of zinc and trimethlysilylchloride, but the application of both of these
methods to (194) resulted in the isolation of starting material only. Soai^^ used
sodium borohydride in mixed alcohol solvents and this method was also applied to
(194); the product was not the expected alcohol (242) but, suprisingly, the
epoxy-alcohol (244) isolated as a single stereoisomer.
It would appear that at reflux a small stationary concentration of the monocyclic
tautomer (242) was present which provided a means for the total reduction of (194),
with hydride attacking from the least sterically hindered face of the ketone. At first
sight this reaction is unusual but becomes less so when it is realised that the ketone
carbonyl is reactive towards borohydride but the carbamate carbonyl is not and that
for all measurable tautomeric equilibria of homotropan-l-ols, the proportion of
monocyclic tautomer increases with an increase in temperature. Although there was
no spectral evidence for the presence of (243) at room temperature, it is not
unreasonable to expect a small proportion of epoxy-ketone at elevated temperature
and hence a conversion into (244). This raises the question as to why the N-methyl
epoxide (196) was not similarly reduced, as this had withstood hydride in refluxing
THF. In this case lithium aluminium hydride had been used which reacted more
rapidly with the carbamate moiety of (194) than with the small amount of (243). The
resulting amine (196) presumably existed completely in the bicyclic form even at
67°C, and was therefore resistant to attack by hydride.
The reduction of (243) to (244) was reversible: the alcohol could be oxidised
86
back to (243) using Jones reagent, which confirmed the structure as an epoxy-ketone.
The CIS-1,4 stereochemistry of (244) was elucidated from the NMR spectrum
which showed a doublet of doublets at 5 2.91 (J = 4.5,3.6 Hz) and a triplet at 8 3.11 (J
= approximately 4.8 Hz) for the epoxide protons. The larger coupling of 4.5 Hz
corresponds to a vicinal coupling between these protons, leaving two similar
couplings of 3.6 Hz and approximately 4.8 Hz to the a-nitrogen and a-hydroxy
protons. Had a trans-\,A relationship existed, two significantly different values would
have been observed and such an effect will be seen later for trans-\,A systems. The
conversion of (244) into the exo-epoxide (235) is outlined in figure 5.6.
HNR HNR RN
UCl/DMSO _ , NaH/THF/ ^
DM&W*
OR' c i(246) (235)
"B uLlrtsa I R = C O jC H iP h
K* L*. (245) R'=OTs Figure 5.6
The alcohol (244) was tosylated to (245) in 92% yield and then converted into
the chloride (246), as a single stereoisomer as shown by NMR spectroscopy. The ^H
NMR spectrum of (246) showed the epoxide protons as a doublet of doublets at 6 3.14
and a triplet at 5 3.37, with a mutual vicinal coupling of 4.3 Hz. A coupling of
approximately 4.8 Hz was seen between the epoxide and a-nitrogen protons, but the
corresponding coupling to the a-chloride proton was much larger (J = 8.9 Hz),
confirming the 1,4-franj stereochemistry. Cyclisation of (246) into (235) was
achieved in near quantitative yield using sodium hydride as base and the product had
identical spectroscopic properties to a sample prepared by oxidation of the alkene
(238).
87
5.3 SYNTHESIS OF EiVDO-6,7-EPOXYHOMOTROPANE
In 1957 Henbest^*^ reported that epoxidation of cycIohex-2-enol (247) with
peroxybenzoic acid afforded the cw-epoxide (248) selectively (figure 5.7). The
directing effect of the hydroxy group has been developed into a highly efficient
method of constructing versatile intermediates in organic synthesis.
— o80%
Figure 5.7
Henbest attributed this steering effect to hydrogen-bonding between the
hydroxy group and 0(2) of the peroxyacid (figure 5.8.A). Whitham^*^ investigated
the syn-stereodirecting effect in more detail, studying the preferred transition state
geometry of the process. Finally, Sharpless"^ proposed that the hydroxy group is
co-ordinated to 0 (1) of the peroxyacid whose remaining lone pair then became
favourably aligned with the it-system of the double bond (figure 5.8.B). Koôovsky^
reported a similar pronounced syn-stereodirecting effect for allylic carbamate groups.
ArAr
O
Figure 5.8BFigure 5.8A
An anomolous result was obtained by Cope in medium ring allylic alcohols.
The stereoselectivity of peroxyacid epoxidation of cyclooct-2-enol (249) was reversed
88
giving the trans-epoxide (251) only,** and cyclohept-2-enol (252) gave a mixture of
cis- and tronj-epoxides**® (253) and (254) respectively. These observations were
rationalised by Teranishi“ after extensive epoxidation studies of five- to
nine-membered ring allylic alcohols with both MCPBA and transition metal-catalysed
peroxyacids. Interestingly, the use of vanadium-catalysed epoxidation afforded the
cM-epoxides (250) and (253) with very high syn-stereoselectivity (figure 5.9).
OH
(249) (250)
Monoperoxyphthalic acid : <1 : >99 (81%)VO(acac)2/SuOOH: 97 : 3 (83%)
OH OH(253) (254)
Monoperoxyphthalic acid : 66 : 34 (95%)VO(acac)/BuOOH: >99 : <1 (87%)
Figure 5.9
It was therefore anticipated that the known allylic alcohol (150) would undergo
peroxyacid epoxidation with high anti-stereoselectivity, thus providing an
intermediate which could be converted to the endo-epoxyhomotropane (259).
Reaction of (150) with 1.2 equivalents of MCPBA did indeed furnish the
trans-epoxide (255) in 95% yield as a single stereoisomer (figure 5.10). The *H NMR
spectrum of (255) was similar to that of the cw-epoxide (244) prepared previously and
showed a similar mutual vicinal coupling of 4.7 Hz between epoxide protons.
However, two larger couplings of 8.2 Hz and 9.3 Hz to a-nitrogen and a-hydroxy
protons were observed, compared to respective values of 3.6 Hz and approximately
89
HNR HNR HNR
MCPBA 1
93*
Jones reagent
96*
OH OH(150) (255)
HNR
(256)
NaH/THF
93*
(259)OR'
"BuLi/TsCll (^ ^ ) R=H92* I—». (258) R'=OTs
R = C02CH2Ph
Figure 5.10
4.8 Hz for (244), confirming the /rans-relationship of the epoxide and a-substituents.
The crude *H NMR spectrum showed no signals that might correspond to the
cis-epoxide.
The hydroxy substituent of (255) was inverted to give the trans-lA isomer
firstly by oxidation with Jones reagent to (256). All spectroscopic data pointed to the
epoxy-ketone existing as the monocyclic tautomer only at room temperature with, for
example, a characteristic carbonyl signal at 8 206.2 in the NMR spectrum. This is
in sharp contrast to the epoxy-ketone (194) with the opposite epoxide stereochemistry,
which was discovered previously to exist as the bicyclic tautomer only. Reduction of
(256) with sodium borohydride then afforded the epimeric alcohol (257) as a single
stereoisomer. The remaining two steps were straightforward; tosylation of (257) gave
(258) which cyclised with base to give the endo-epoxyhomotropane (259) in excellent
yield. The ^H NMR spectrum of (259) differed from that of the exo-epoxide (235)
with the chemical shift of the epoxide protons appearing 0.53 ppm further downfield
and a coupling of 3.6 Hz between epoxide and bridgehead protons.
90
5.4 REDUCTION OF N-PROTECTED EPOXYHOMOTROFANES
The final steps necessary to prepare die parent 6,7-epoxy-homotropane and
norhomotropane skeleta were hydride reduction and hydrogenolysis respectively. The
epoxide group of the 1-hydroxy derivative (194) withstood these reaction conditions
and it was hoped that similar chemoselectivity could be achieved for the exo-epoxide
(235). Reduction of (235) with lithium aluminium hydride in THF afforded the
N-methyl epoxide (260) and hydrogenolysis using a palladium catalyst gave the
corresponding norhomotropane (261), with the epoxide surviving both reductions as
anticipated (figure 5.11). The N-methyl amine was stored as the hydrochloride salt
although the NMR spectra were recorded on the free amine.
MeN HCl
O
UAIH^THF/HO,
86%(235)
(260)
HN
Figure 5.11
(261)
Reduction of the N-alkoxycarbonyl group simplified the NMR spectra and the
two amines were easily identifiable. The *H spectrum of (260) showed a singlet at 8
2.55 for the newly formed methyl group and a second singlet at 8 3.43 for the two
equivalent epoxide protons. Five signals were observed in the spectrum including
a methyl carbon signal at 8 46.5. The spectra of (261) were similar, again showing no
coupling between the epoxide and the bridgehead protons in the *H NMR spectrum.
91
The enJo-epoxide (259) was also subjected to hydride and hydrogenolysis
reductions, but in contrast to (235) the epoxide moiety was considerably more
reactive being reduced to give homotropan-7a-ol (262) and norhomotropan-7a-ol
(263) under normal conditions. However, the N-methyl epoxide (264) could be
isolated if milder reducing conditions were employed (figure 5.12), although the
product was contaminated with 10% of the alcohol (262).
MeN
COgCHzPhN
70*
o (259)
H2Œd/C
HO (262)
UAlH4/-78°OHa54*
.HO
HO
Figure 5.12
(263)
O (264)
The ^H NMR spectrum of (262) displayed a doublet of triplets at 6 4.55 for the
a-hydroxy proton, due to vicinal coupling to both Cg protons (J = 10.5, 6.4 Hz) and
the Cg-bridgehead proton (J = 6.4 Hz). The NMR spectrum showed methyl and
methine (CHOH) carbon signals at 6 39.2 and 6 71.7 respectively. The epoxide
protons of (264) were coupled to the bridgehead protons (J = 3.2 Hz) in the *H NMR
spectrum confirming the endo- (rather than exo-) stereochemistry of the epoxide. A
parallel reactivity difference between homologous exo- and endo-epoxytropanes is
noted in the next chapter and a discussion of their relative reactivities will follow
92
later.
5.5 CONCLUSION
In section 5.2 it was demonstrated that cjco-epoxidation of a protected
homotrop-7-ene (238) proceeded in good yield with high stereoselectivity, in accord
with the behaviour of bridgehead functionalised analogues. However, the difficulties
encountered in the synthesis of precursors limited the practical application of this
approach, although preliminary investigation into the facial selectivity of the double
bond was very promising. The potential to prepare dihydroxylated derivatives of
homotropane using osmium chemistry^ is a possibility here.
A more efficient approach was then developed which allowed the introduction
of the epoxide function at an earlier stage of the synthesis, providing the
exo-homotropane (235) via a serendipitous hydride reduction. The reversal of
stereoselectivity for cyclooct-2-enols from syn- to anti- was used to full advantage to
prepare the enJo-epoxyhomotropane (259). After simple reductions, a full range of
novel oxygenated homotropane derivatives was then isolated which are believed to be
the first such derivatives to be reported.
It was concluded that the success of this strategy, which uses cheap and
commercially available reagents for all transformations, could easily be adapted to the
tropane series and may ultimately lead to successful synthesis of tropane natural
products.
93
CHAPTER 6
Stereoselective Oxygenation of Tropane
6.1 INTRODUCTION
Tropane derivatives having epoxy and hydroxy groups in the 2-carbon bridge
are of considerable natural importance.*’ Alkaloids such as scopolamine (based on
the exo-6,7-epoxytropane structure), bao gong teng (8) and the calystegines (198 -
200) all possess Cg-oxygenation and have been discussed previously. A further
example is provided by the schizanthine class of tropane alkaloid which are all diester
derivatives of 3a,6P-dihydroxytropane. These compounds were isolated in 1987 from
Schimnthus grahmif^^ and Schizanthus pinnatus}^^ There are ten such alkaloids in
total and Schizanthine F (265) is illustrated here (figure 6.1).
MeN
Schizanthine F (265)
Figure 6.1H
MeMe
Me
In view of the wealth of natural products containing Cg,C^ functionality the
application to tropanes of the epoxidation strategy already developed for
homotropanes (chapter 5) was considered to be a worthwhile investigation.
6.2 SYNTHESIS OF EJrO-6,7-EPOXYTROFANE
The method of choice for preparing exo- and endo-epoxyhomotropanes
involved epoxidation of a monocyclic alkene prior to cyclisation to the homotropane,
rather than epoxidation of an N-protected homotrop-7-ene. The key precursor for this
synthesis is therefore the previously prepared allylic alcohol (227). Cope had
94
reported that epoxidation of cyclooct-2-enol with peroxyacid proceeded with high
anti-stereoselectivity,**^ whereas treatment of cyclohept-2-enol under similar
conditions afforded a mixture of cis- and trans- epoxides.**® A similar outcome was
expected for the cyclohept-2-enol derivative (227) and this was indeed the case;
treatment with 1.2 equivalents of MCPBA gave (266) and (267) in a respective ratio
of 62:38. At this stage it was realised that if an approach to exo- and
endo-epoxytropanes was to be developed, the overall efficiency of the two syntheses
could be incre^ed by varying the ratio of isomers isolated from the epoxidation
reaction. Reagents other than MCPBA were therefore explored and the results are
summarised in figure 6.2. All yields quoted are isolated pure yields and the ratios are
calculated from *H NMR integrations of crude reaction mixtures.
HNCOoCHnPh
OH(227)
HNCO,CH,Ph
O
OH(266)
HNCOoCHoPh
O
OH(267)
Reagent RatioYield
MCPBA/CH^Oz
MMPP/EtOWHzO
VO(acac)2/muOOH
62
42
>95
Figure 6.2
38
58
<5
93%
72%
<40%
Magnesium monoperphthalate*^* (MMPP), a recently developed peroxygen
product, was the first alternative reagent to MCPBA chosen. Using a literature
protocol*^ (225) was epoxidised in a hydroxylic solvent (necessary to dissolve the
reagent) and a slight shift towards the frww-epoxide (267), compared to MCPBA,
resulted. A vanadium-catalysed peroxyacid procedure**^ was also used to epoxidise
(227) and, in good agreement with literature precedent,**^ a high syn-stereoselectivity
95
was observed but unfortunately the yield was disappointing. The high yielding
MCPBA epoxidation, along with the easy separation of the two epoxides by
chromatography meant that this was the most efficient means of obtaining (266) and
(267).
The stereochemistiy of each of the two epoxides was elucidated after
comparison of the *H NMR spectra with those of the higher homologues (244) and
(255). The tran^^-epoxide (267) had a mutual vicinal coupling of 5.0 Hz between
epoxide protons and further couplings of 5.8 and 6.6 Hz to the a-nitrogen and
a-hydroxy protons. The homologous tranj-epoxide (255) showed similar, slightly
larger couplings of 8.2 and 9.3 Hz to a-nitrogen and a-hydroxy protons. A similar
pattern was observed for the cw-epoxide (266) which displayed no coupling between
the epoxide, a-nitrogen and a-hydroxy protons in contrast to respective small
couplings of 3.6 and 4.8 Hz for the homologous cw-epoxide (244). These assignments
were supported by the chemical evidence of high syn-stereoselectivity for the
vanadium-catalysed epoxidation and, ultimately, further spectral evidence from
comparison of exo- and entio-epoxytropanes with homotropane homologues.
The exo-epoxide in the homotropane series was introduced by epoxidation of
the tautomeric mixture of a,p-unsaturated ketone (151) and hemiaminal (152) as
described in chapter 5.2. No such tautomerism existed for the a,p-unsaturated ketone
(229) and as a result this was unreactive to MCPBA (figure 6.3). However, treatment
of (229) with alkaline hydrogen peroxide afforded the cw-epoxy ketone (269) as the
only product in 69% yield. This suggests that tautomerism of (229) is pH dependent
and, in the presence of base, a small concentration of the electrophilic hemiaminal
(230) was present which was readily epoxidised by MCPBA. The epoxy-ketone (269)
isolated from this reaction existed only in the monocyclic form with no spectral
evidence for the hemiaminal (268).
96
HNR
NaOH/HgOg.ôM,
MCPBA
RN
OH
(230)
R = C02CH2Ph
Figure 6 3
(269)
The epoxy-ketone (269) was also synthesised by Jones oxidation of the
epoxy-alcohol (266) in 87% yield (figure 6.4). The NMR spectra of (269) prepared
by each of these methods were identical. The NMR spectrum displayed a
characteristic carbonyl carbon signal at S 208.5. The ^H NMR spectrum showed two
epoxide signals at 8 3.37 and 5 3.47 with only a small vicinal coupling of 0.9 Hz
observed between the epoxide proton and the a-nitrogen proton.
Jones reagent
HNCOnCHoPh
O
OH (266)
87*
Figure 6.4
HNCOoCH,Ph
O
O
Two methods for introducing the trans-1,4 stereochemistry necessary for
cyclisation were now possible. The epoxy-alcohol (266) could be tosylated and the
tosylate replaced by chloride with inversion, or the epoxy-ketone (269) could be
reduced with sodium borohydride and tosylated. The former route was chosen and
97
the synthesis of the exo-epoxy tropane (272) is outlined in figure 6.5. Both the
formation of the tosylate (270) and subsequent transformation into the chloride (271)
proceeded in excellent yield. Cyclisation was then achieved by treatment of (271)
with sodium hydride and the NMR spectrum of (272) was similar to that of the
exo-epoxyhomotropane (235). As a result of slow N-CO rotation, two doublets at Ô
3.42 (J = 3.2 Hz) and 8 3.45 (J = 3.2 Hz) were observed for the epoxide protons
together with two broad singlets at 8 4.33 and 8 4.41 for the bridgehead protons. The
absence of measurable coupling between these two sets of signals confirmed the
exo-stereochemistry of the epoxide group. Reductive deprotection of (272) is
reported in section 6.4.
HNCOjCHoPh
O
OR
HNCOoCH^
NaH/THFoDME, 87%
Cl
(272)”BuLi/Tsa (266) R -H
99% [_^ (270)R = OTs Figure 6.5
6.3 SYNTHESIS OF £M )0-6,7-EP0XYTR0FANE
Starting from the franj-epoxide (267), the identical approach to that described
above was applied to the formation of the endo-epoxide (275). The alcohol (267) was
tosylated to (273) and then converted into the chloride (274) as a single stereoisomer
(figure 6.6). Both of these transformations proceeded smoothly but attempte to
cyclise (274) using a variety of bases including sodium hydride and potassium
carbonate failed.
98
HNCCkCHoPh
O
OR
HNCOnCHoPb
O
"BuLi/rsCl92%
(267)R = H
(273)R = OTs
(274)
Figure 6.6
(275)
A better leaving group, namely a tosylate, was required to induce cyclisation
and this was prepared by firstly oxidising the alcohol (267) with Jones reagent and
then reducing the ketone (276) with hydride (figure 6.7). The epoxy-ketone (276)
existed only as the monocyclic tautomer; the NMR spectra were well resolved at
room temperature and no evidence of the broadening associated with tautomerism
was observed. The NMR spectrum displayed a carbonyl signal at ô 209.3
HNR HNR
Jones reagent
OH(267)
R = C02CH2Ph
(276)
[H]
HNR HNR
Figure 6.7
Reagent
NaBH/THF/A:
L-Selectride/THF/-780C:
O
OH (267)
O
OH (277)Ratio
62
45
38
55
Yield
(95%)
(89%)
99
assigned to the ring-carbon Cj. The homologous 8-membered ring epoxy-ketone
(256) gave only the trans-1,4 isomer (257) on reduction with sodium borohydride, but
unfortunately reduction of (276) with borohydride gave a mixture of both cis-1,4
(267) and trans-1,4 (277) isomers in a ratio of 62:38 (calculated from NMR
integrations) which were separable by chromatography. L-Selectride^^ was also used
to reduce (276) and this proved to be a superior reagent, giving a richer mixture of the
trans-1,4 (277) in good overall yield. The redundant cw-1,4 alcohol (267) could be
recycled by oxidation to the ketone (276) followed by reduction. The stereochemistry
of (277) was confirmed by the small vicinal coupling of 1.6 Hz between the epoxide
and a-hydroxy protons in the NMR spectrum.
Successful formation of (275) was then achieved by tosylation and treatment of
the tosylate (278) with sodium hydride in THFzDME solvent (figure 6.8).
HNCO^CHiPh
O
OH
NaH/THF/oDME, 61%
OTs
(277)Figure 6.8
A more elegant method of preparing the trans-1,4 alcohol (277) would involve
direct inversion of the alcohol group (rather than oxidising and reducing from the
opposite face) and several literature methods for achieving this are known. The
epoxy-alcohol (255), based on the 8-membered ring, was more readily available than
(267) and this was used for preliminary investigations. Kellogg*^ inverted the
stereochemistry of secondary alcohols by first mesylating the alcohol, inverting the
mesylate with cesium propionoate in DMF solvent and then hydrolysing the resulting
ester to the alcohol. This method was reported to be remarkably clean, proceeding
without detectable amounts of elimination or racémisation products. The mesylate of
(255) was duly prepared but when (279) was subjected to the conditions of Kellogg, it
100
o
OH
NbO/EtgN97% ^
HNCOoCHoPh
O
OMs
(255) (279)
HNCOoCHnPh
Ms = MeSO'O
OH (257)
Figure 6.9
failed to give the desired inversion product; only starting material was isolated (figure
6.9). Dcegami reported a modification of this procedure, inverting the
stereochemistry of a mesylate with cesium acetate using benzene as solvent in the
presence of 18-crown-6. Applying this method to (279) resulted in no observable
inversion of the mesylate. When the reaction was repeated at higher temperature only
a complex mixture of products was isolated.
Corey inverted the mesylates of secondary alcohols using potassium
superoxide in DMSO solvent, but this also failed when applied to (279). Finally,
direct inversion of the alcohol (255) was tried using the Mitsunobu^^ procedure but
without success. The failure of these reactions was disappointing but nonetheless, the
overall efficiency of the oxidation and reduction method was acceptable.
101
6.4 REDUCTION OF N-PROTECTED EPOXYTROPANES
The epoxide moiety of N-alkoxycarbonyl-protected epoxyhomotropanes was
able to withstand both hydride and hydrogenolysis reduction and as a result, both
exo-epoxyhomotropane (260) and enrfo-epoxyhomotropane (264) were accessible. A
similar reductive procedure was applied to the N-protected exo-epoxytropane (272),
but the epoxide was reactive and was ring-opened by DIB AH at 0°C to give
tropan-6p-ol (282) (figure 6.10). However, when the reduction was repeated at -78°C,
COzCHgPhN
MeN HCl
(272)
DIBAH/-78°C/
DIBAHAfcy HO. 70*
M%#*
H2mdKV(280)H
N HO
MeN
HO
Figure 6.10
the carbamate was reduced chemoselectively to afford 6p,7p-epoxytropane (280).
The epoxide of (272) was also stable to hydrogenolysis; reduction at 1 atmosphere
with a palladium catalyst gave 6p,7P-epoxynortropane (281). Both (280) and (281)
were volatile and were handled as the hydrochloride salts.
The emfo-epoxide (275) was subjected to reduction with lithium aluminium
hydride in refluxing diethyl ether and both epoxide and carbamate moitiés were
reduced to give tropan-6a-ol (283) (figure 6.11). An attempt to reduce the carbamate
group selectively by repeating the reaction at -78°C led only to the isolation of (283).
102
The hydrogenolysis of (275) was not attempted but ring-opening of the epoxide to the
alcohol would have been expected in accord with behaviour of the homologous
endo-epoxyhomotropane (259).
COzCHzPhN
UA1H /Et20/A ►
69*
(275)Figure 6.11
The substantially greater reactivity of endo-epoxides compared to exo-epoxides
in both the homotropane and tropane ring systems is surprising. The preferred
conformation of the seven-membered ring of homotropane is known to be the boat
form*^^ and this gives no reason to expect significant steric shielding of the
eWo-face. However, the established role of bridging nitrogen in the hydride
reduction of a C=C bridge in 7-azanorbomenes*^° suggests that the nitrogen may
assist the attack of hydride reducing agents on the epoxide from the exo-face,
explaining the increased lability of the eWo-epoxide to hydride reducing agents.
103
CHAPTER 7
Total Synthesis of Scopine, Pseudoscopine & Nor-derivatives
7.1 INTRODUCTION
The first synthesis of scopine (6), the precursor to scopolamine and
(-)-hyoscine, was reported by Fodor®" in 1959. This synthesis was discussed in
chapter 2.5 in connection with epoxidation studies of the homotrop-7-ene system.
The efficiency of the epoxidation step was low because of the difficulties associated
with epoxidation of an alkene in the presence of an amine group. Despite the
extensive interest in the synthesis of other natural products based on azabicyclic
systems (for example, the numerous syntheses of anatoxin-a" "' ), endeavours to
design an effective synthesis of scopine were neglected until an elegant procedure
was reported by BackvalP^ in 1991. Based on earlier work involving the preparation
of simpler tropane alkaloids^® (chapter 1.3), Backvall prepared both scopine and
pseudoscopine using palladium-catalysed 1,4-chloroacetoxylation of a functionalised
cycloheptadiene in the key step of the synthesis. The synthesis of scopine (6) is
summarised in figure 7.1. Treatment of (284) with palladium(II) acetate, lithium
chloride, lithium acetate and benzoquinone afforded the 1,4,6-trisubstituted
cycloheptane (285) which was then reduced to the alcohol (286). A requirement for
preparing 3 a-hydroxy isomers of tropane alkaloids {i.e scopine) is a tran.s-relationship
between the nitrogen and benzyloxy groups. This was achieved by treating (286) with
sodium p-toluenesulphonamide in the presence of a palladium(O) catalyst, the reaction
proceeding with retention of configuration. The hydroxy group of (287) was inverted
into the chloride (288) and then epoxidised to (289) with MCPBA. The epoxidation
step gave high syn-stereoselectivity (>98%). Subsequent cyclisation of (289),
induced with potassium carbonate, produced the 8-azabicyclo[3.2.1]octane structure.
The N-methyl group was introduced by removing the N-tosyl group of (290) using a
dissolving metal reduction followed by quenching with methyl iodide to give scopine
benzyl ether (291). Hydrogenolysis then gave scopine (6) in 8 overall steps.
Scopoline (7) was obtained as a by-product in the final deprotection step. The overall
104
yield for these transformations from the diene (284) to scopine was 16%.
Cl HNTs
oP(KOAc)2/
o cH Ph ^Benzoquinone,
63*(284)
OCH2PhNaNHTsÆ>d(0),
66*
OH
DIB AH
96*
HNTs
(1)Na(2) Mel
88*
PhCHgO
- R = Ts(290)
-► R = Me(291)
Figure 7.1
HVPdA:
R = Ac(285)
R = H (286)
" OCHzPh
OCHzPh
(287)
Mzd/ua
HNTs
MCPBA
86*
Cl
OCHzPh
(288)
HO
Scopine (6)
Scopoline (7)
HO O
Backvall then applied this strategy to prepare the 6p~analogue of scopine,
namely pseudoscopine (297), as outlined in figure 7.2. A stereochemical requirement
for the synthesis of (297) is a cw-relationship between the nitrogen and the benzyloxy
group in the monocyclic precursors. The conversion of (285) into (292) therefore
required retention of configuration of the OR group which was achieved by treatment
with sodium p-toluenesulphonamide in the absence of a palladium catalyst. The
acetate group of (292) was saponified and the resultant alcohol was protected as the
silyl ether (293). The bulky silyl ether group facilitated selective syn-epoxidation
giving (294) as the major epoxide in a ratio of 87:13. Deprotection of the silyl ether
and mesylation of the alcohol afforded (295) which was cyclised to (296) using
potassium carbonate as base. The final deprotection and méthylation steps were
105
identical to those described for the synthesis of scopine. The overall yield for the
conversion of the diene (284) into pseudoscopine (297) was 25% and the relatively
high efficiency of these two syntheses makes them the first effective preparations of
scopine and pseudoscopine.
HNTs
OCHzPh79*
OAc (2S5)
(1) NaOH. 84*
(2)TBDMSC1.94*
MeN
- OCHzPh
OR
- R = OAc (292)
. R = TBDMS(293)TsN
MCPBA
84*
HNTs
OCHjPh
OTBDMS
(1)TBAF(2)M,OÆ(3N
steps
77*overall
60*
Pseudoscopine (297) (296)OH
TBDMSCl = ‘BuMe2 SiCl
HNTs
O
OMs
TBAP-Bu^NP Figure 7.2
The reaction between pyrroles and oxyallyls is an established method of
producing the tropane system and is significant as it allows access to uncommon
trop-6-ene analogues. There are drawbacks to this approach; firstly both the pyrrole
and the oxyallyl precursors often have to be highly substituted for the cycloaddition to
proceed in a satisfactory yield. Secondly, the known methods of generating oxyallyls
involve hazardous chemicals. In 1992, Mann^' reported a notable improvement to
this approach by generating oxyallyls from polybromoketones and diethylzinc. Using
this method, a range of both nitrogen- and oxygen-bridged bicycles could be
synthesised safely and in high yield. In addition, N-alkoxycarbonyl-protected
pyrroles readily intercepted the oxyallyl intermediates and this has a specific bearing
on the synthesis of epoxytropane derivatives as the nitrogen is suitably protected for
106
subsequent epoxidation. Mann demonstrated this with a synthesis of scopoline (7)
which could be produced in gramme quantities (figure 7.3)
I^^^NCOgMe
Figure 7.3
Br
Br — / Br
MeN
COoMeN
O(DagZn(2)Zn/Cu
59*(298)
O
HO
Scopoline (7)
COgMeN
DIB AH53*
Reaction of tetrabromoacetone and methyl pyrrole-N-carboxylate in the
presence of diethylzinc followed by debromination of the cycloadduct produced (298)
in good yield. Epoxidation of (298) with MCPBA then gave the exo-epoxide (299)
which was reduced with an excess of DIB AH to afford scopoline (7). The attack of
hydride presumably occurs exclusively from the P-face of the ketone as no
by-products were reported for the reaction. In view of the high stability observed for
exo-epoxytropanes (chapter 6) the use of milder reducing agents, or indeed a reversal
in the order of epoxidation and reduction steps, may well diversify the range of
products available from (298). This intermediate offers the potential to prepare
scopine and pseudoscopine and this may result in future elaboration of the procedure.
The shortage of approaches to scopine warranted an investigation into a further
synthesis. To this end, a new route to scopine, pseudoscopine and novel
nor-derivatives is described here and is complimentary to the method reported by
Backvall.
107
7.2 A STRATEGY TO SCOPINE AND PSEUDOSCOPINE
PRECURSORS
An efficient method of preparing exo-epoxytropanes was discussed in the
previous chapter and this forms the basis for the synthesis of 3-oxygenated
derivatives. The additional 3-hydroxy substituent could be introduced by performing
the initial nitroso cycloaddition reaction using cyclohepta-1,3-dienol (M2). This
diene could could be prepared from tropone*^* but this approach was rejected in view
of the difficulties encountered by Howarth^^ in producing (302) in high yield, and also
because tropone is not a readily-available starting material. Instead, the procedure of
Schiess and Wisson*^^ was used which involved the peroxyacetic epoxidation of
cycloheptatriene (300) and reduction of the resulting epoxide (301) with lithium
aluminium hydride (figure 7.4). For large scale preparation of (302), the epoxide was
not purified prior to reduction. The overall yield was low but all the reagents were
commercially available and of low cost. The reaction could also be scaled up to
produce gramme quantities of the dienol.
r v .CHgCOgH/NazCOg ^ ^ UAIH4
} L y(301) (302)
Figure 7.4
The stereoselective introduction of the 3-hydroxy group was of prime
importance for a successful synthesis of scopine (3a-hydroxy) and pseudoscopine
(3p-hydroxy). When (302) was subjected to the standard nitroso cycloaddition
conditions an acceptable yield of cycloadducts was produced but the reaction showed
little selectivity giving a 35:65 mixture of (303) and (304) which could not be
separated by chromatography (figure 7.5). The stereostructures of (303) and (304)
were assigned on the basis of the relative chemical shifts of the a-hydroxy protons.
Two such signals were observed in the NMR spectrum with a difference in
chemical shift of greater than 5 0.5. The signal attributed to the 3a-proton of the
108
C ^ “(302)
+73%
HO Hp OHHa(303)
Figure 7.535 65
major cycloadduct (304) was notably up-field (5 3.67) as this lay within the shielding
cone of the double bond. The 3p-proton of the minor adduct (303) was observed
much further down-field at 5 4.24. These shifts are comparable to values for similar
adducts (bearing different substituents at nitrogen) prepared by Howarth.®^
At this stage of the synthesis two options were available: either the mixture of
adducts could be used in subsequent steps, with the hope of separation at a later stage,
or attempts could be made to prepare purer samples of (303) and (304). The latter
approach was chosen as it was anticipated that difficulties might arise during attempts
to identify and characterise compounds from reaction mixtures.
The stereoselective reduction of tropinone (10) has been extensively studied by
Beckett^^^ who reported the participation of kinetic and thermodynamic factors in the
course of the reduction. Noyori^^ reported similar effects in the reduction of
6,7-dehydrotropinone. The structure of tropinone is similar to that of the adducts
(303) and (304) and by analogy the reduction of the ketone, formed by oxidation of
these alcohols, would be expected to furnish the 3a-hydroxy stereoisomer (303) as the
major component (kinetic control). It was therefore decided to oxidise the alcohols
first and then reduce (figure 7.6). Oxidation of the cycloadducts with Jones reagent
produced the ketone (305) in high yield. The product was identified from an
additional carbonyl signal at 5 206.0 in the NMR spectrum and by the absence of
a-hydroxy protons in the Ô 3.0 - 5.0 region of the *H NMR spectrum. Stereoselective
reduction of the ketone under conditions of kinetic control was achieved by treatment
109
(303) : (304) 35 : 65 OH
Jones reaagent
90*
(305)O
R = C02CH2Ph
TBDMSa = %uMe2SiaL-Selectride,
85*
TBDMSO
(306)
TBDMSa
imidazole91*
Figure 7.6
HO(3a-0H)
(303)85
(304)15
of (305) at -78°C with L-Selectride^^ to afford an 85% yield of alcohols. Inspection
of the *H NMR spectrum of the crude reaction product indicated a substantial
enrichment (compared to the cycloaddition reaction) of the 3a-hydroxy isomer (:M)3)
which comprised 85% of the mixture. This was deemed to be of high enough purity
for subsequent reactions.
The hydroxy group of (303) was protected in order to allow manipulation of
other groups at a later stage of the synthesis. The versatile r-butyldimethlysilyl
protecting group was chosen since this was reported* to be sufficiently stable to
withstand a wide range of reaction conditions and yet is easily introduced and
removed. The alcohol (303) was protected as the silyl ether (306) using standard
conditions. The *H NMR spectrum of (306) confirmed a successful reaction showing
diagnostic up-field methyl signals; this compound was a suitably protected and
functionalised precursor to scopine. The 3a-silyl ether (306) did however contain a
minor impurity, namely the 3P-silyl ether, formed from the epimeric alcohol (304).
110
These stereoisomers could not be separated at this stage of the synthesis but, in
subsequent transformations, products derived from this mixture could be isolated in
pure form.
It was considered initially that a possible method of generating an enriched
sample of the 3P-silyl ether (308) would be to invert the stereochemistry of the
3a-hydroxy group of (303) and then protect as the silyl ether. However, because of
the difficulties encountered previously with such reactions (chapter 6.3) this approach
was rejected. Instead, it was decided to investigate the effect of silylating
cyclohepta-3,5-dienol (302) with TBDMS-Cl prior to the cycloaddition reaction, in
the hope that a bulky group would alter the stereoselectivity. The silyl ether (307)
was prepared in 93% yield and then reacted with benzyl nitrosoformate (figure 7.7).
COzCHgPh
TBDMSd/imidazole,
93*
+
20OTBDMSTBDMSO
(306)
R=C02CH2Ph
- R = H(302)
► R = TBDMS(307)
Figure 7.7
This approach was successful; from the integrations in the *H NMR spectrum of
the reaction mixture a 20:80 ratio of the adducts (306):(308) had been formed in
favour of the 3p-silyl ether (308). The a-silyl ether proton of (308) was observed in
the NMR spectrum as an approximate triplet of triplets (J = 10.3, 6.3 Hz) at 5 3.68.
The chemical shift of this proton was notably up-field as a result of shielding by the
double bond. This observation is in good agreement with the corresponding value for
the a-hydroxy proton of (304) which had a very similar shift (6 3.67). The minor
signals in the NMR spectra of (308) corresponded to the other epimer (306).
I l l
7.3 SYNTHESIS OF SCOPINE AND NORSCOPINE
The silyl ether (306) was converted into the protected gxo-6,7-epoxytropane
system (314) using an identical sequence of reactions to those described for the
preparation of (272). These transformations are outlined in figure 7.8.
HNR
HNR
MŒBA
TBDMSO
OTBDMS
OH (309)
O OTBDMS
OH
92*4*
HNR
R = C02CH2Ph
(311)O OTBDMS
OR'
RN
TBDMSO
LiCl/DMSO'BuLi/TsCl85*
HNR
R' = OTs (312)O •• OTBDMS
Cl Figure 7.8
NaH/THF
DME.82*
) (313)
Reduction of (306) with sodium amalgam afforded the alcohol (309) in good
yield which displayed an exchangeable hydroxy proton at 5 2.47 in the *H NMR
spectrum. Epoxidation of (309) produced a pair of stereoisomers (310) and (311) in a
3:7 ratio based upon isolated yields of the epoxides after chromatographic separation.
The epoxidation reaction had a moderate stereoselectivity in favour of the product
(311) having the epoxide tram to the silyl ether and thus the correct stereochemistry
for subsequent conversion into the cxo-epoxytropane (314). The ^H NMR spectra of
(310) and (311) were broad at ambient temperature but the vicinal couplings of the
112
epoxide protons were discernible; assignment of stereochemistry was made on the
basis of the magnitude of these values, in conjunction with the coupling constants of
(266) and (267). The stereoisomer (311) having the epoxide cis to the a-N and a-O
displayed two doublets at ô 3.23 and § 3.32 due to the epoxide protons with a mutual
vicinal coupling of 5.0 Hz. The corresponding signals of (310), a triplet at ô 3.13 (J =
7.5 Hz) and a multiplet at ô 3.18, had additional vicinal couplings to the a-0 and a-N
protons. The trans-1,4 stereochemistry between nitrogen and a leaving group
(necessary for cyclisation) was achieved by tosylating (311) and treating the tosylate
(312) with lithium chloride to give (313). Chromatographic purification of the
tosylate intermediate (312) removed minor by-products that had arisen from
6p-silylether contamination of (306). The inversion of configuration in the
chlorination step was confirmed from a vicinal coupling of 4.6 Hz between epoxide
and a-chlorine protons in the ^H NMR spectrum of (313). Nucleophilic displacement
initiated by sodium hydride as base then gave the doubly-protected tropane structure
(314). As a result of slow rotation about the N-CO bond, the epoxide and bridgehead
signals were duplicated in the ^H and ^ C NMR of (314). The epoxide protons
appeared as doublets in the ^H NMR spectrum at 8 3.53 and 8 3.56, with a mutual
coupling of 3.5 Hz. The absence of any observable coupling between epoxide and
bridgehead protons confirmed the exo-epoxy stereochemistry of (314).
The final deprotection steps are shown in figure 7.9. Reduction of (314) with
lithium aluminium hydride in diethyl ether produced (315) in high yield which
showed a singlet at 8 2.51 in the ^H NMR spectrum and a methyl signal at 8 42.5 in
the NMR spectrum, assigned to the N-methyl group. As anticipated from the
behaviour of (272) (the analogue of (314) without a 3-oxy substituent) the epoxide
was stable to the reducing conditions. Removal of the O-protecting group of (315)
was achieved using TBAF^^^ in THF solvent to furnish scopine (6), the spectroscopic
properties of which were in good agreement with literature data. A discussion of the
^H NMR spectmm of scopine will follow a description of the synthesis of
pseudoscopine.
113
COgCHzPhN
LîAlH
95*
TBDMSO (314)
TBAF81*
TBAF
81*
TBDMSO (315) HOScopine (6)
COnCHoPhN
HN
(316)
H2/PdA:100*
HO
Figure 7.9
Norecopine (317)
HO
Hydrogenoiysis of (314) instead of hydride reduction, with subsequent removal
of the O-protecting group from (316), allowed for the preparation of norscopine (317)
a non-natural derivative of scopine. A partial synthesis of norscopolamine (from
scopolamine) has been reported, but no other total synthesis of norscopine has been
undertaken to date.
7.4 SYNTHESIS OF PSEUDOSCOPINE AND NORPSEUDOSCOPINE
A partial synthesis of pseudoscopine was reported by Heusner and Ziele*^^ in
1958 and involved oxidation of scopine and subsequent reduction with potassium
borohydride. The yield for the latter step was low and this was not an effective
synthesis. The method of preparing (297), reported by Backvall, therefore represents
the only practical synthesis of pseudoscopine to date. A second synthesis is described
here which involves an identical procedure to that described for scopine but using the
SP-silyl ether (308) as starting material in the initial step (figure 7.10).
114
HNR
R
OTBDMS
OTBDMS
MCPBA
80*
O OTBDMS
OH
+
HNR
R = C02CH2Ph
(320)
O OTBDMS
OR'
RN
NaH/THF
DME,
(323)OTBDMS
HNR
O OTBDMS
(322)
"BuLi/T#a84*
R' = H(320)
R' = 0Ts(321)
Figure 7.10
Reduction of (308) with sodium amalgam produced the allylic alcohol (318)
which was epoxidised with MCPBA to furnish a 1:1 mixture of (319):(3M). The
epoxides were separable by chromatography; the isomer (320) with the epoxide cis to
the a-N and a-O was isolated in pure form but (319) contained minor impurities
resulting from contamination by the 6a-silyl ether. The epoxidation reaction was
unselective and although alternative epoxidising agents could have been used to vary
the ratio, (319) is a potential precursor to analogues of scopine (chapter 7.6) and is
therefore a synthetically useful by-product which was retained. The cw-epoxide (320)
was tosylated to (321) and then substituted with chloride ion to give (322).
Cyclisation of (322) using sodium hydride gave the protected form of pseudoscopine
(323).
The deprotection procedures used to prepare pseudoscopine (297) and
norpseudoscopine (326) were the same as those applied to (314) and are outlined in
115
figure 7.11. Hydride reduction of (323) gave the N-methyl derivative (324) which
was treated with fluoride ion to give pseudoscopine (297). Alternatively, initial
desilylation of (323) followed by hydrogenoiysis reduction of (325) furnished
norpseudoscopine (326), a novel analogue.
COzCHzPhN
MeN
MeN
(323) OTBDMS
TBAF
83*
OTBDMS
TBAF
OH
Pseudoscopine (297)
COzCHoPhN
HN
Figure 7.11
Norpseudoscopine (326)
(325) OHOH
The successful synthesis of scopine and pseudoscopine was confirmed after
comparison of *H, and mass spectra with published literature data^^ which
showed a good match. The vicinal couplings between the a-hydroxy and
neighbouring protons differed substantially for the two isomers in the ^H NMR
spectra (figure 7.12). The 3P-hydroxy proton of scopine (6) at 5 4.20 is in an
equatorial position and only coupling to each of the proximate axial protons was
observed resulting in a triplet. Vicinal coupling to equatorial protons was too small to
be measured which is typical of equatorial-equatorial coupling. The 3a-proton of
pseudoscopine (297) at Ô 4.16 is in an axial position and as a consequence the vicinal
axial-axial couplings are much larger and the signal appeared as a triplet of triplets (J
= 9.7, 6.7 Hz). The nor derivatives (317) and (326) showed similar values. This
116
MeN
Scopine (6)HO
Pseudoscopine(297) OH
J e q ^ = 5 . 3 H z J » x . a x - 9 / 7 H z
J ^ «;=O H z J«.cq = 6.7Hz
Figure 7.12
effect was also observed for the 3a- and SP-oxygenated adducts, such as (303) and
(304), where the Cg-proton coupled to pseudo-axial and pseudo-equatorial protons
and confirmed the earlier assignments based on chemical shift.
7.5 FURTHER FUNCnONALISATION OF TROPANE
This chapter has described a successful synthetic strategy to scopine,
pseudoscopine and novel nor-derivatives. The anticipated stability of the exo-epoxide
group has been used to full advantage in this synthesis. During the course of this
work, two by-products (310) and (319) were isolated which had the epoxide group
trans to the a-O and a-N. These epoxides are potential precursors to ‘emfo-scopine’
(327) and ‘emfo-pseudoscopine’ (328) as depicted in figure 7.13. There is no
reference to such compounds in the literature to date. As the endo-6,7-epoxide of the
tropane system has twen shown to be significantly more labile to hydride reduction
than the exo-epoxide, a different protecting group at nitrogen would be required at the
conclusion of the synthesis.
117
HNR
O - OTBDMS
OH (310)
MeN
Steps►
HO
HNR
O
OH (319)
Endo-scopine (327)
MeN
OTBDMSSteps
►
Figure 7.13
OHEndo-pseudoscopine (328)
A preliminary investigation into introducing oxygen functionality in the Cg and
C4 locations of tropane was studied. This could provide further analogues of scopine
and would also be of use in preparing other oxygenated tropane alkaloids. The
epoxide (301) was an intermediate in the synthesis of cyclohepta-3,5-dienol (302) and
it was reasoned that the diene in (301) might undergo a cycloaddition reaction with
nitroso compounds. A small portion of (301) was purified and subjected to the
standard cycloaddition conditions employed previously. The reaction was successful
and two isomeric epoxides (329) and (330) were isolated in an 80:20 ratio (Figure
7.14). The ^H NMR spectrum displayed characteristic bridgehead protons of similar
R
Q ‘(301)
C02CH2Ph
64*
Figure 7.14
+ ISOMER
(330)
80:20
118
shift to those of cycloadducts prepared previously. The two isomers could not be
separated by chromatography and the assignment of relative stereo- and
regiochemistries from NMR signals in the spectrum of the mixture could not be
made with confidence. However, recrystallisation of the product mixture afforded the
major cycloadduct (329) in pure form and a crystal was grown for X-ray analysis
(appendix 1). This revealed that (329) had the stucture depicted in figure 7.14 with an
endo-epoxide group.
Although the stmcture of (330) could not be elucidated without more extensive
spectroscopic study, this reaction does provide a highly functionalised cycloadduct
(329) in pure form and in an acceptable yield. Using this as a starting material, there
is a clear scope to produce further natural and non-natural derivatives of tropane.
119
APPENDIX 1
X-Ray Crystal Data for the Epoxide Cycloadduct (329)
X-RAY CRYSTAL DATA FOR THE EPOXIDE CYCLOADDUCT (329)
Crystals suitable for X-ray single crystal determination were obtained from a 1 ml solution containing 20 mg of the recrystallised cycloadduct (329) in diethyl ether
which was slowly evaporated at 0°C. The crystal used for data collection was a colourless block (0.48 x 0.45 x 0.39 mm) mounted on a glass fibre.
Ciystal data for C15H15NO4: M = 273.3, orthorhombic space group Pna2i, a = 16,711 (4), b = 9.688 (2), ç = 16.316 (2) Â, V = 2641.5 (9) Z = 8, = 0.10 mm-\Mmo-koc) = 0.7107 Â, F(000) = 1152, = 1.374 mg/M^.
The unit cell parameters were determined from the optimised setting angles of 22 automatically centered reflections, 9.9 < 20 < 23.0°. The intensities of 2452 reflections were measured on a Siemens P4 diffractometer (5 < 20 < 46°) using a w scan technique. 3 Check reflections monitored every 100 reflections indicated no crystal decomposition. The data were conected for Lorentz and polarisation effects and merged to give 2011 unique reflections (Rj t .032) and 1547 reflections with F > 4a(F).
The structure was solved by direct methods and refined by full matrix least squares
using the programme SHELXTL - pc. ^® Two unique but approximately superimposable molecules were found in the asymmetric unit.
Hydrogen atoms were included in calculated positions (C-H = 0.96 Â). The number of atoms refined with anisotropic displacement parameters was limited by the data to parameter ratio.
Final cycles of least-squares refinement used a weighting scheme w = l/(o^F + 0.0007 F^) and wR 0.0078. The final diffence Fourier map was featureless (+0.007 and -0.002e Â'^).
The geometry of the molecule (329) is shown in figure A. 1
120
X-ray Crystal Structure of the Enoxide CyclnaHHuct (329)
010a
C20cC19a O
(329)
Figure A.l
121
T a b le 1 . A tomic c o o r d i n a t e s (xlO ) and e q u i v a l e n t i s o t r o p i c
d i s p l a c e m e n t c o e f f i c i e n t s (Â^xlO^)
X y z U(eq)
C ( l ) 1075(6) 1368(10) 4459(8) 64 (3)C(2) 559(5) 1340(10) 5215(8) 68 (4)C(3) 840(6) 2133(9) 5930(9) 56(3)C(4) 1549(5) 3024(9) 5915(8) 56(3)C(5) 2047(5) 3199(9) 5163(8) 47(2)N(6) 2392(4) 1 8 4 2 (6) 4971(7) 57(3)0 (7 ) 1857(3) 838(5) 4664 53 (2)C(8) 1 1 2 6 (8) 2833(13) 4092(9) 80(5)C(9) 1623(7) 3724(10) 4474(9) 72(5)0 (1 0 ) 7 5 6 (4) 3652(6) 5901(7) 71(3)C ( l l ) 3154(5) 1 6 6 8 (8) 4768(7) 41(3)0 (1 2 ) 3661(3) 2548(6) 4867(6) 57(2)0 (1 3 ) 3311(3) 408(5) 4493(6) 50(2)C(14) 4162(5) 47(9) 4416(9) 62 (4)0 (1 5 ) 4 2 0 2 (4) -14 5 1 (4 ) 4238(6) 47 (2)0 (1 6 ) 3862 -2381 4790 58 (2)0(17) 3892 -3794 4631 75(3)0 (1 8 ) 4261 -4278 3920 70(3)0(19) 4601 -3349 3368 73 (3)0 (2 0 ) 4571 -1935 3527 62 (3)0(1A) 3445(5) 3 7 0 3 (9) 2077(8) 52(2)0(2A) 2918(5) 3730(8) 1341(8) 56(4)C(3A) 3162(5) 2800(9) 644(8) 51(3)0(4A) 3834(5) 1829(9) 675(8) 57(4)C(5A) 4321(5) 1684(9) 1444(8) 56(3)N ( S A ) 4 7 4 2 (4) 2967(6) 1601(8) 52 (3)0(7A) 4253 (3) 4094(5) 1835(6) 65(3)C(8A) 3426(7) 2358 (14) 2 5 0 6 (9) 80(5)0(9A) 3852(7) 1343(10) 2181(9) 70(4)O(IOA) 3014 (3) 1325(6) 702(6) 63(2)C ( l lA ) 5511(5) 3 0 0 3 (8) 1874(7) 38(3)0(12A) 5958 (3) 2045(5) 1827(6) 53(2)0(13A) 5697(3) 4257(5) 2127(6) 53(2)0(14A) 6 5 4 8 (4) 4518(8) 2249(8) 53(3)C(ISA) 6644 (3) 6036 (4) 2377(5) 4 0 ( f )C(16A) 7077 6518 3050 50(2)0(17A) 7151 7934 3186 62(3)0(18A) 6793 8868 2648 57(2)0(19A) 6360 8386 1976 62(3)C(20A) 6285 6970 1840 53 (2)
* E q u i v a l e n t i s o t r o p i c tJ d e f i n e d a s one t h i r d o f t h e t r a c e o f t h e o r t h o g o n a l i z e d t e n s o r
122
Table 2. Bond lengths (Â)
C ( l ) -C (2) 1 .5 0 7 (17) 0 ( 1 ) - 0 ( 7 ) 1 .4 4 4 (11)C ( l ) - C ( 8 ) 1 .5 4 2 (16) C (2 ) -C (3 ) 1 .4 7 3 (18)C(3) -C (4) 1 .4 6 6 (12) 0 ( 3 ) - 0 ( 1 0 ) 1 .4 7 9 (10)C(4) -C(5) 1 .4 9 2 (17) 0 ( 4 ) - 0 ( 1 0 ) 1 .4 5 9 (10)C(5) -N{6) 1 .4 6 9 (11) 0 ( 5 ) - 0 ( 9 ) 1 .4 2 2 (18)N ( 6 ) - 0 ( 7 ) 1 .4 1 3 (8) N(6) -0 (1 1 ) 1 .3 2 7 (11)C (8 ) -C (9 ) 1 .3 5 0 (17) 0 ( 1 1 ) - 0 ( 1 2 ) 1 .2 1 2 (10)C ( l l ) -0 (1 3 ) 1 .3 2 7 (10) 0 ( 1 3 ) - 0 ( 1 4 ) 1 .4 6 9 (9)C ( 1 4 ) -0 ( 1 5 ) 1 .4 8 1 (10) 0(1A) -0(2A) 1 .4 8 8 (16)C (1A)-0(7A) 1 .4 5 8 (10) 0(1A) -0(8A) 1 .4 8 0 (17)C(2A) -C(3A) 1 .5 0 8 (16) 0(3A) -0(4A) 1 .4 6 6 (12)C(3A) -O(IOA) 1 .453 (10) C(4A) -C(5A) 1 .5 0 2 (17)C(4A) -O(IOA) 1 .4 5 6 (10) 0(5A) -N(6A) 1 .4 5 0 (11)C(5A) -C(9A) 1 .4 7 3 (18) N(6A) -0(7A) 1 .4 1 6 (9)N(6A) - C ( l lA ) 1 .3 6 0 (11) 0(8A) -0(9A) 1 .3 2 6 (17)C ( l lA ) -0(12A) 1 .1 9 5 (10) O (l lA ) - 0 ( 1 3 A) 1 .3 2 0 (10)0(13A) -C(14A) 1 .4 5 9 (9) 0(14A) -0(15A) 1 .4 9 5 (9)
T a b l e 3 . Bond a n g l e s
0 (2 ) - 0 ( 1 ) - 0 ( 7 ) 1 0 8 .8 (9 ) . C ( 2 ) - C ( l ) - C ( 8 ) 1 1 1 .5 9)0 ( 7 ) - 0 ( 1 ) - 0 ( 8 ) 1 1 1 .6 (8 ) 0 ( 1 ) - 0 ( 2 ) - 0 ( 3 ) 1 1 7 .2 8)0 ( 2 ) - 0 ( 3 ) - 0 ( 4 ) 1 2 3 .4 (1 1 ) 0 ( 2 ) - 0 ( 3 ) - 0 ( 1 0 ) 1 1 7 .7 10)0 (4 ) - 0 ( 3 ) -0 (1 0 ) 5 9 .4 (5 ) 0 ( 3 ) - 0 ( 4 ) - 0 ( 5 ) 1 2 2 .2 11)0 ( 3 ) - 0 ( 4 ) - 0 ( 1 0 ) 6 0 .7 (5 ) 0 ( 5 ) - 0 ( 4 ) - 0 ( 1 0 ) 1 1 6 .6 10)0 (4 ) - 0 ( 5 ) -N(6) 1 0 7 .0 (8 ) 0 ( 4 ) - 0 ( 5 ) - 0 ( 9 ) 1 1 4 .3 9)N(6) - 0 ( 5 ) -0 ( 9 ) 1 1 0 .3 (1 0 ) C ( 5 ) - N ( 6 ) - 0 ( 7 ) 1 1 6 .3 6)0 ( 5 ) - N ( 6 ) -0 (1 1 ) 1 2 2 .9 (7 ) 0 ( 7 ) - N ( 6 ) -0 (1 1 ) 1 1 5 .5 7)0 (1 ) - 0 ( 7 ) -N(6) 1 1 4 .2 (6 ) 0 ( 1 ) - 0 ( 8 ) - 0 ( 9 ) 116 .3 11)0 (5 ) - 0 ( 9 ) -0 ( 8 ) 1 1 6 .3 (1 0 ) 0 ( 3 ) - 0 ( 1 0 ) - 0 ( 4 ) 5 9 .9 5)N ( 6 ) - 0 ( 1 1 ) - 0 ( 1 2 ) 1 2 3 .2 (8 ) N ( 6 ) - 0 ( 1 1 ) - 0 ( 1 3 ) 1 1 3 .0 7)0 ( 1 2 ) - 0 ( 1 1 ) - 0 ( 1 3 ) 1 2 3 .6 (7 ) 0 ( 1 1 ) - 0 ( 1 3 ) - 0 ( 1 4 ) 1 1 6 .0 6)0 ( 1 3 ) - 0 ( 1 4 ) - 0 (1 5 ) 1 0 7 .1 (6 ) 0 ( 1 4 ) - 0 ( 1 5 ) -0 (1 6 ) 1 1 9 .2 6)0 ( 1 4 ) - 0 ( 1 5 ) - 0 ( 2 0 ) 1 2 0 .8 (6 ) 0(2A) -O(IA) -0(7A) 1 0 9 .0 9)0 (2 A ) -O (IA )-0 (8 A ) 1 1 2 .6 (8 ) 0(7A) -O(IA) -0(8A) 1 1 2 .1 7)0(1A) -0(2A) - 0 ( 3 A) 1 1 6 .0 (7 ) C(2A) -0(3A) -0(4A) 1 2 4 .4 11)0 ( 2 A) -0(3A) -O(IOA) 1 1 9 .5 (1 0 ) 0(4A) -0(3A) -O(IOA) 5 9 .8 5)0(3A) -0(4A) -0(5A) 1 2 0 .2 (1 0 ) C(3A) -0(4A) -O(IOA) 5 9 .7 5)0(5A) -0(4A) -O(IOA) 1 1 7 .0 (1 0 ) 0(4A) -0(5A) -N(6A) 1 0 9 .2 9)0(4A) -0(5A) -0(9A) 1 1 4 .5 (9 ) N(6A) -0(5A) -0(9A) 1 0 7 .9 10)0(5A) -N(6A) -0(7A) 1 1 5 .4 (6 ) 0(5A) -N(6A) -O ( l lA ) 1 2 2 .5 7)0(7A) -N(6A) -O ( l lA ) 1 1 5 .9 (7 ) 0(1A) -0(7A) -N(6A) 1 1 4 .1 5)0(1A) -0(8A) -0(9A) 1 1 6 .9 (1 2 ) C(5A) -0(9A) -0(8A) 1 1 6 .5 10)0(3A) -O(IOA) -0(4A) 6 0 .5 (5 ) N(6A) -O ( l lA ) -0(12A) 1 2 3 .4 8)N(6A) -O ( l lA ) -0(13A) 1 1 0 .4 (7 ) 0(12A) -O ( l lA ) -0(13A) 1 2 5 .9 8)0 ( l lA ) - 0 (13A) -0 (14A) 1 1 5 .6 (6 ) 0 (13A )-0 (14A ) -0(15A) 1 0 7 .1 6)0(14A) -0(15A) -0(16A) 1 1 9 .7 (5 ) 0(14A) -0(15A) -O(20A) 1 2 0 .3 5)
123
T a b l e 4 . A n i s o t r o p i c d i s p l a c e m e n t c o e f f i c i e n t s (Â^xlO^)
U , , U , u u TJ11 22 33 12 13 23C(2) 37(5) 5 6(6) 111(10) -2 (5 ) - 4 ( 6 ) -2 (6C(3) 68(6) 3 7(5) 64 (7) 0(5) 4(6 ) 0(5C(4) 53(5) SO (5) 64(7) 9(5) -1 2 (6 ) - 1 7 ( 5N(6) 35(4) 3 1(4) 1 0 4 (7) -4 (3 ) 0 (5 ) -26 (40 (7 ) 29(3 ) 3 9(3) 92 (5) 4(3) -6 (4) - 1 4 ( 4C(8) 92(9) 114(9 ) 35(7 ) 41(8) - 3 (6 ) 7 (6C O ) 89(8) 50(6) 78 (9) -6 (6) 30(7) - 1 1 ( 60 (1 0 ) 67(4 ) 6 1(4) 86(5) 13 (3) 16 (4) -14 (4C ( l l ) 34(5) 40(5) 48 (6) -6 (4) - 1 ( 5 ) -12 (50 (1 2 ) 43 (3) 57(3) 71(5) -12 (3 ) 7(4) - 1 7 ( 40 (1 3 ) 29(3 ) 45(3) 75(5) 2(3) -2 ( 3 ) - 1 6 (3C(14) 35(4) 59(6) 93(8) 6(5) - 7 (5 ) -13 (6C(2A) 58 (6) 36(5) 75(7) -4 (4 ) 8(6) -8 (5C(3A) 53 (5) 43(5 ) 56(6) -9 (5 ) -4 (6) 1 (5C(4A) 54(6) 50(5) 68(7 ) -3 (5 ) 7(6) - 7 ( 5N(6A) 35(4) 2 5(4) 97(7) 2(3) - 1 2 (4) -25 (40(7A) 26 (3) 3 4(3) 133(7) 2(2) - 6 (4 ) -23 (4C(8A) 76(7) 110(9) 53 (8) -19(7) -1 3 (7 ) 2 (7C(9A) 88(8) 45(5 ) 78 (9) -9 (5 ) -3 6 (7 ) 12 (6O(IOA) 48(3) 52(4) 90(5) -4 (3 ) - 9 (4 ) - 1 7 ( 4C ( l lA ) 33 (4) 44(5 ) 38 (5) -2 (4 ) - 1 ( 5 ) -3 (50(12A) 3 5(3) 43(3 ) 82(6) 12(3) 1(4) -7 (40(13A) 2 5(3) 39(3) 96 (5) -5 (2 ) -7 ( 3 ) -22 (3C(14A) 33 (5) 55(5) 71(7) -5 (4 ) -5 ( 5 ) - 8 ( 5
The a n i s o t r o p i c d i s p l a c e m e n t e jq jo n en t t a k e s t h e form :
-2, ^ (h^a*^n11
+ 2hka*b*a )
124
T a b le S . H-Atom c o o r d i n a t e s (xlO ) and i s o t r o p i c
d i s p l a c e m e n t c o e f f i c i e n t s (Â^xlO^)
X y z U
H(1A) 837 774 4056 80H(2A) 509 390 5376 80H(2B) 35 1663 5070 80H(3A) 715 1727 6452 80H(4A) 1830 3136 6424 80H (SA) 2482 3813 5285 80H (SA) 802 3096 3632 80H(9A) 1705 4631 4253 80H(14A) 4445 261 4912 80H(14B) 4390 567 3973 80H(16A) 3608 -2047 5279 80H(17A) 3658 -4435 5012 80H (ISA) 4281 -5252 3811 ' 80H(19A) 4855 -3682 2879 80E(20A) 4805 -1294 3147 80H(IAA) 3255 4400 2447 80H(2AA) 2882 4663 1147 80H(2AB) 2392 3450 1510 80H(3AA) 3051 3170 110 80H(4AA) 4119 1660 174 80H(SAA) 4710 966 1366 80H(8AA) 3123 2244 3002 80H(9AA) 3858 447 2432 80H ( 1 4 0 6851 4214 1783 80H(14D) 6736 4032 2725 80H(16B) 7324 5875 3420 80H(17B) 7449 8266 3649 80H(18B) 6844 9842 2742 80H(19B) 6113 9029 1606 80H(20B) 5988 6638 1377 80
125
CHAPTERS
Experimental
INSTRUMENTATION
Routine NMR spectra were recorded on Varian EM 390 (90 MHz) or Jeol JNM-PSlOO (100 MHz) spectrometers. Higher field ^H NMR (300 MHz, 250 MHz) and (75 MHz) NMR spectra were recorded on a Bruker AM 300 or an ARX 250 spectrometer. Chemical shifts were recorded in ppm (6) downfield from the internal reference (TMS). Signal characteristics are described using standard abbreviations: s (singlet), d (doublet), dd (doublet of doublets), t (triplet), q (quartet), quin (quintet), m (multiplet), br (broad) and v (very); protons identified as NH or OH were shown to be exchangeable with D2O. In some circumstances, signals that appear in a more simplified form than the molecule allows are give the prefix For example, a dddd which appears as a quintet is quoted as -quin. Where data are quoted for two tautomers or rotamers, overlapping signals aie shown in italics but may be quoted separately for reasons of clarity even though they are not fully resolved or assigned. In the spectra, C, CH, CH2, CH3 are used to indicate quaternary, methine, methylene and methyl carbons respectively, as shown by off-resonance decoupling or DEPT experiments.
IR spectra were recorded on PE 1604 FT or PE 298 IR spectrometers as solutions in CH2CI2 unless indicated otherwise. Band intensities aie described using standard abbreviations: s (strong), m (medium), w (weak), br (broad), v (very).
Mass spectra were measured routinely on a VG Micromass 14 spectrometer and were obtained using ionisation by electron impact except where chemical ionisation was used (shown Cl) or fast atom bombardment (shown FAB); intensities are given as percentages of the base peak. Accurate mass measurements were obtained using a Kratos Concept mass spectrometer either at Leicester University or through the SERC service at the University College of Swansea.
Melting point measurements were made using a Kofler hot stage apparatus and are
uncorrected.
Combustion Analyses were performed by Butterworth Laboratories Ltd., Teddington, Middlesex.
126
TECHNICAL
Reactions were performed under dry nitrogen using solvents dried by standaid methods. Diethyl ether was dried over sodium wire and distilled from LiAlH^. Dichloromethane, toluene and benzene were distilled from calcium hydride. Petroleum ether and ethyl acetate were distilled prior to use. Methanol and ethanol were purified with magnesium and iod ine .T etrahydrofuran was distilled from sodium-benzophenone. Triethylamine and pyridine were distilled from potassium hydroxide. All other solvents were dried and purified as described by Perrin.
Flash chromatography was caiTied out according to the method of StilE" using Merck Kieselgel 60 (230 - 400 mesh). Thin-layer chromatography was conducted on standard commercial aluminium sheets pre-coated with a 0.2 mm layer of silica gel (Merck 60 - 254).
Tétraméthylammonium periodate^^^Paraperiodic acid (18.75 g, 0.082 mol) was dissolved in water (45 ml) and added in portions to a stirred 25% solution of tétraméthylammonium hydroxide (30.04 g, 0.082 mol) at 0°C. The precipitated white solid was filtered, washed with cold methanol (40 ml) and dried to afford tétraméthylammonium periodate (18.02 g, 83%) as a crystalline white solid. M.p. 252 - 254°C.
Benzvl-N-hydroxvcarbamate^"^^
Benzyl chloroformate (25.1 ml, 0.176 mol) was dripped into a stirred solution of hydroxylamine hydrochloride (13.41 g, 0.19 mol) and sodium hydroxide (15.80 g, 0.40 mol) in water (180 ml) at 0°C. On complete addition the mixture was warmed to room temperature and stirred for 4.5 hr. The pH was adjusted to 2 by adding HCl solution (6M) and the liberated oil was then repeatedly extracted with diethyl ether (3 X 90 ml). The combined organic layers were washed with water (50 ml) and the ethereal layer dried over anhydrous sodium sulphate. After filtration and evaporation of solvent at reduced pressure, the yellow solid was recrystallised twice from toluene and petroleum ether (b.p. 60 - 80°C) to yield benzyl-N-hydroxycarbamate (20.89 g, 71%) as a crystalline white solid. M.p. 67 - 69°C (Lit.^^ m.p 71°C).
N-(Benzyloxvcarbonvl)-7-aza-8-oxabicvclor4.2.21dec-9-ene (103) Tétraméthylammonium periodate (37.90 g, 0.143 mol) and cycloocta-1,3-diene (95)
127
(14.32 g, 0.132 mol) in dichloromethane (180 ml) was stiiTed at 0°C. A solution of benzyl-N-hydroxycarbamate (23.90 g, 0.143 mol) in dichloromethane (85 ml) was dripped in over 20 min. The mixture was warmed to room temperature on complete
addition and stirred for a further 5 hr. The solution was filtered, washed with saturated sodium thiosulphate solution (3 x 70 ml) and water (50 ml). The organic layer was separated, dried over anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure. The residual yellow oil was purified by flash chromatography using 1:9 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford (103) (22.07 g, 61%) as a white solid. M.p. 60 - 61 °C (Lit^» m.p. 61.0 - 61.5°C).
(90 MHz, CDCI3): 1.45 - 2.30 (series of m, 8H), 4.65 (m, IH, a-N), 4.90 (m, IH, a-O) 5.15 (s, 2H, CH^Ph), 5.75 (dd, J =10, 7 Hz, IH), 6.30 (dd, J = 10, 7 Hz, IH), 7.30 (s, 5H).
Potassium azodicarboxvlate^"^
Azodicarbonamide (30.0 g, 0.26 mol) was stirred with a 1:1 solution of potassium hydroxide (75 ml) at 0°C. After the complete evolution of ammonia, the solution was filtered leaving a yellow solid. This was dissolved in the minimum quantity of cold
water (80 ml) and poured into 4 volumes of ethanol, resulting in the precipitation of a bright yellow solid, which was filtered and washed with methanol (3 x 25 ml). The process was repeated to afford potassium azodicarboxylate (24.1 g, 48%), which was dried over P2O5.
N-(Benzvloxvcarbonvl)-7-aza-8-oxabicvclor4.2.21decane (131)
Into a stirred slurry of potassium azodicarboxylate (17.68 g, 0.09 mol) and (103) (5.03 g, 0.018 mol) in methanol (155 ml) at 0°C was added dropwise glacial acetic acid (10.5 ml, 0.18 mol) over 40 min. The mixture was allowed to warm to room temperature and stilled for 3 hr when a further portion of potassium azodicaiboxylate (7.28 g, 0.038 mol) was added and stirred for a further 2 hr. Water (12 ml) was added and the bulk of the solvent removed under vacuum. The oily residue was partitioned between dichloromethane (105 ml) and 5% sodium bicarbonate solution (40 ml). After washing with further bicarbonate solution (2 x 40 ml) the organic layer was dried over magnesium sulphate. Filtration and removal of solvent under vacuum gave
a crude oil which was purified by flash chromatography using 2:8 diethyl ether:petroleum ether (b.p. 40-60°C) to yield (131) (4.59 g, 91%) as a colourless oil.
128
ÔH (300 MHz, CDCI3): 1.51-1.79 (series of m, lOH), 1.99 (brm, IH), 2.13 - 2.28 (brm, IH), 4.48 (brm, IH, a-N), 4.59 (brm, IH, a-O), 5.21 (s, 2H, CHzPh), 7.28 - 7.37 (m, 5H).
8(2 (75 MHz, CDCI3), Signais in italics were broadened due to rotation about the N-CO bond and values are approximate; the signal due to Cj of the benzyl group was not visible; 20.5, 22.3, 23.7, 24.6, 33.0, 34.4 (6 x CHj), 51.2 (CBN), 67.1 (CH2Ph),76.5 (CHO), 127.8 (2 x aryl CH), 128.3 (aryl CH).
v^ax (CH2CI2): 3090W, 3060w, 3040w, 2930s, 2860m, 1725s, 1690s, 1590w, 1495w, 1400m, 1355m, 1330m, 1315m, 1290m, 1260m, 1210w, 1195m, 1150w, 1095s, 1070s, 1020m, 990w, 930w cm'^.
™/z (%): 275 (M+, 13), 231 (7), 146 (5), 140 (7), 132 (10), 92 (14), 91 (100), 81 (4).
C16H21NO3 [M+] requires ™/z 275.1521; observed 275.1525.
Ciy-4-r(Benzvloxvcarbonvl)aminolcycIooctanol (132)To a solution of (131) (1.26 g, 4.56 mmol) in dry ethanol (18 ml) was added sodium phosphate (3.12 g, 22 mmol). The suspension was stirred for 5 min and freshly prepared®^ and powdered 6% sodium amalgam (14 g) was added at 0°C under a nitrogen atmosphere. After stirring for 3 hr the solution was filtered and the solvent removed under vacuum. The oil was dissolved in water (40 ml) and extracted into dichloromethane (3 x 20 ml). The combined organic layers were dried over anhydrous magnesium sulphate, filtered, and the solvent removed under vacuum to afford (132) (1.26 g, 100%) as a colourless oil.
ÔH (300 MHz, CDCI3); 1.48 - 1.74 (brm, 12H), 2.15 (brs, OH), 3.65 (brm, IH, a-N),
3.81 (brm, IH, a-O), 4.94 (d, J = 7.3 Hz, IH, NH), 5.06 (s, 2H, CH2Ph), 7.26 - 7.37 (m, 5H).
8c (75 MHz, CDCI3): 22.0, 23.4, 28.0, 30.9, 31.2 & 33.2 (6 x CH2), 51.2 (CHN), 66.5 (CH2Ph), 71.2 (CHO), 128.0 (2 x aryl CH), 128.4, (aryl CH), 136.6 (aryl C), 155.4 (C=0).
Vmax (CH2CI2): 3600m, 3420m, 3015w, 2930s, 2860m, 1710s, 1500s, 1450w,
129
1315brw, 1215s, lOSOw, 1055s, 1010m, 975m, 910m cm \
“ /z (%): 277 (M+, 6), 168 (19), 146 (34), 142 (31), 126 (16), 125 (21), 124 (17), 123 (19), 114 (10), 112 (23), 110 (13), 109 (18), 108 (100), 107 (100), 105 (14).
C16H23NO3 [M+] requires ^ /z 277.1680; observed 277.168.
4- F(B enzvloxvcarbonvD aminolcyclooctanone ( 133)Jones reagent,® prepared form chromium trioxide (12.35 g), concentrated sulphuric acid (11.5 ml) and water (20 ml), was dripped into a solution of (132) (1.234 g, 4.45 mmol) in dry acetone (36 ml). A persistent orange colouration indicated complete oxidation and excess oxidant was destroyed by dropwise addition of isopropanol. The mixture was filtered through celite and the bulk of the solvent was distilled under reduced pressure. The oil was dissolved in dichloromethane (30 ml) and washed with water ( 2 x 7 ml) and brine (5 ml). The organic layer was separated, dried over magnesium sulphate, filtered, and the solvent removed under vacuum to yield (133) (1.206 g, 98%) which, after recrystallisation from petroleum ether (b.p. 80 - 100°C), had m.p. 117 - 118°C. On leaving a solution of (133) to stand in CDCI3 for several hours a second set of minor signals were observed which corresponded to the bicyclic tautomer (134) in a ratio of 71:29. Data for the bicyclic tautomer (134) are given separately below; the figures quoted in italics are common to both tautomers.
5jj (300 MHz, CDCI3), Monocyclic tautomer; 1.33 - 1.65 (series of m, 4H), 1.82 -2.04 (series of m, 2H), 2.07 - 2.31 (brm, 2H), 2.34 - 2.50 (brm, 3H), 3.75 (m, IH), 4.96 (brd, J = 6.9 Hz, NH), 5.07 (s, 2H), 7.34 (s, 5H).
ÔC (75 MHz, CDCI3); 23.4, 28.2, 28.4, 31.2, 39.8 & 40.4 (6 x CHg), 50.8 (CHN), 66.5 (CH^Ph), 128.0 (2 X aryl CH), 128.4, (aryl CH), 136.5 (aryl C), 155.4 (NC=0), 217.0 (CC=0).
The bicyclic tautomer (134) showed signals as shown in italics above together with signals at 8 4.32 (m, IH), 5.11 & 5.17 (ABq, J = 12.4 Hz, 2H, CH^Ph) in the ^H NMR spectrum and at 8 23.0, 23.7, 27.8, 34.7, 37.7 & 40.0 (6 x CHg), 55.4 (CHN), 66.6 (CH^Ph), 92.9 (COH), 127.7, 128.0 & 128.5 (3 x aryl CH), 136.4 (aryl C), 155.1 (C=0) in the ^^C NMR spectrum.
130
V m a x (CH2CI2): 3440m, 3340brw, 3060w, 3015w, 2940s, 2860m, 1720vs, 1505s, 1465m, 1450m, 1405w, 1340m, 1310m, 1215w, 1150w, 1125w, 1085m, 1060w, 1035m, 1025w, 1005w, 980w, 845w cm '\
"i/z (%): 275 (M+, 10), 184 (6), 146 (7), 140 (15), 108 (9), 92 (9), 91 (100), 84 (14).
Found: C, 69.93; H, 7.46; N, 5.09%.C16H21NO3 requires: C, 69.79; H, 7.69; N, 5.10%.
4-Methvlene-r(benzvloxvcarbonvl)amino1cvclooctane (135)
A 50 ml two-necked round bottom flask was charged with sodium hydride (80% dispersion in mineral oil, 183 mg, 3.82 mmol). The flask was equipped with a rubber septum cap and condenser. The system was alternately purged and evacuated with nitrogen; diy DMSO (4 ml) was injected via syringe and the system was warmed to 80°C with stirring for 45 min. On cooling to room temperature methyltripheny- phosphonium bromide (1.363 g, 3.82 mmol) in DMSO (5 ml) was introduced and stirred at room temperature for 10 min. A solution of (133) (300 mg, 1.09 mmol) in DMSO (4 ml) was injected and stirred at room temperature for 18 hr. The bulk of the solvent was distilled under reduced pressure, the residue dissolved in water (15 ml) and extracted with dichloromethane (3 x 20 ml). The combined organic layers were dried over anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure. After flash chromatography using 1:10 diethyl ether:petroleum ether (b.p. 40 - 60°C), (135) (142 mg, 48%) was isolated as a colourless oil.
ÔH (300 MHz, CDCI3): 1.49-1.81 (brm, 7H), 1.95 (m, IH), 2.14 - 2.23 (m, 3H), 2.34 (m, IH), 3.73 (m, IH, a-N), 4.78 and 4.80 (each s, IH, =CH2), 4.88 (d, J = 6.8 Hz, NH), 5.06 (s, 2H), 7.25 - 7.34 (m, 5H).
ÔC (75 MHz, CDCI3): 23.4, 30.2, 30.4, 32.0, 32.5 & 33.6 (6 x CH2), 51.1 (CHN), 66.4 (CH2Ph), 111.6 (=CH2), 128.0, 128.1 & 128.4 (3 x aryl CH), 136.6 (aryl C), 150.9 (C=CH2), 155.4 (C=0).
Vmax (C H 2 C I2 ): 3 3 3 0 m , 30 6 5 W , 3 0 3 0 w , 2 9 2 5 s , 2 8 5 0 m , 1 6 9 5 s , 1 6 4 0 w , 1 5 3 0 s , 1 4 8 5 0 m ,
1 405W , 1 3 1 5 m , 1 2 3 5 s , 1 1 4 5 w , llOOw, 1 0 7 5 w , 1 0 3 5 w , 9 7 0 w , 8 8 5 m , 7 4 0 m , 6 9 5 m
131
% (%, CI); 274 (MH+, 100), 230 (52), 140 (20), 138 (24), 123 (22), 108 (32), 91 (73).
C17H24NO2 [MH+] requires ™/z 274.1807; observed 274.181.
N-(Benzvloxvcarbonvl)-l-methyl-9-azabicvclor4.2.11nonane (136)Mercuiy(n) trifluoroacetate (152 mg, 0.36 mmol) was added to a stirred solution of (135) (92 mg, 0.34 mmol) in dry acetonitrile (12 ml) at room temperature. Stirring was continued for 2.5 hr and the bulk of the solvent was distilled under reduced pressure to afford an oil which was dissolved in THF (10 ml). The solution was
cooled to -78°C and sodium borohydride (26 mg, 0.66 mmol) was added with stirring under an nitrogen atmosphere, resulting in the precipitation of mercury. The solution was warmed to room temperature and stirred for a further 30 min. Excess hydride was destroyed by the dropwise addition of water and the solution was dried with
anhydrous sodium sulphate and filtered through celite. The solvent was distilled under reduced pressure and the crude oil was purified by flash chromatography using
1:4 diethyl ether:petroleum ether (b.p. 40 - 60°C) to give (136) (57 mg, 61%) as an oil. When the reaction was repeated under a nitrogen atmosphere, the yield of (136) was increased to 73%. Figures quoted in italics are common to both rotamers.
ôjj (300 MHz, CDCI3), Major rotamer: 1.33 -1.56 (series of m, 6H), 1.60 (s, 3H), 1.80 - 2.15 (series of m, 6H), 4.34 (brt, J = 7.2 Hz, IH), 5.03 & 5.14 (ABq, J = 12.5 Hz, 2H, CH2Ph), 7.25 - 7.37 (m, 5H).
Minor rotamer: 1.33 - 1.56 (series of m, 6H), 1.60 (s, 3H), 1.80 - 2,15 (series of m, 6H), 4.41 (m, IH), 5.03 & 5.14 (ABq, J = 12.5 Hz, 2H, CH2Ph), 7.25 - 7.37 (m, 5H).
Ôq (75 MHz, CDCI3), Major rotamer: 23.6 & 25.0 (2 x CH2), 28.6 (CH3), 29.4, 34.8, 38.8 & 39.6 (4 x CH2), 57.7 (CHN), 63.9 (C), 66.0 (CH2), 127.7,127.8 & 128.4 (3 x aryl CH), 137.2 (aryl C), 154.3 (C=0).
Minor rotamer: 23.5 & 25.0 (2 x CH2), 29.7 (CH3), 29.4, 33.1, 40.4 & 41.2 (4 x CH2),58.7 (CHN), 63.1 (C), 66.6 (CH2), 127.7,127.8 & 128.4 (3 x aryl CH), 137.2 (aryl C),154.3 (C = 0).
Vmax (CH2CI2): 2930s, 2850m, 2690w, 1700s, 1450m, 1440m, 1340m, 1320m, 1120m,
132
1060m cm"\
"i/z (%, Cl): 274 (MH+, 22), 173 (10), 172 (14). 138 (29), 91 (100).
C17H24NO2 [MH"*"] requires "'/z 274.1807; observed 274.181.
N-(Benzvloxvcarbonvl)-l-hvdroxvmethvl-9-azabicvclof4.2.nnonane (137)The hydroxy methyl compound (137) (25 mg, 25%) was isolated as a by-product during chromatographic separation of a crude sample of (136) prepared as described above.
ÔH (300 MHz, CDCI3): 1.26 - 1.66 (series of m, 7H), 1.81 (m, 2H), 2.08 (m, 2H), 2.31 (m, IH), 3.60 & 3.70 (ABq, J = 11.7 Hz, 2H), 4.38 (m, IH), 5.08 & 5.16 (ABq, J =12.4 Hz, 2H), 7.28 - 7.36 (m, 5H).
ÔC (75 MHz, CDCI3): 24.0, 24.1, 29.1, 33.9, 34.2 & 35.8 (6 x CH2 ), 58.6 (CHN), 67.0 (CH2Ph), 68.7 (C), 68.9 (CH2OH), 127.8, 128.0 & 128.5 (3 x aryl CH), 136.6 (aryl C), 156.0 (C=0).
Vmax (CH2CI2): 3385brs, 3075w, 3060w, 3030w, 2930s, 2850m, 1675s, 1545w, 1455m, 1410m, 1345m, 1325m, 1280m, 1233m, 1205m, 1165m, 1120m, 1080m, 1060m, 1020m, 980w, 965w, 910w, 770m cm'^
% (%): 289 (M+, 5), 181 (49), 154 (19), 153 (65), 152 (52), 138 (11), 137 (11), 136(11), 125 (15), 124 (24), 122 (12), 110 (10), 109 (61), 108 (100), 107 (87), 106 (12), 105 (14), 95 (100) 91 (100).
C17H23NO2 [M" ] requires ”*/z 289.1678; observed 289.168.
N-Methvl-l-methvl-9-azabicvclor4.2.11nonane (141)A solution of (136) (102 mg, 0.37 mmol) in dry THF (4 ml) was added dropwise to a stirred sluiiy of LAH (28 mg, 0.73 mmol) in dry THF (1 ml) under nitrogen at 0°C. The system was allowed to warm to ambient temperature, stirred for 3 hr, then heated at reflux for 1 hr. The excess hydride was destroyed by addition of water-saturated diethyl ether and the solution was dried with sodium sulphate and filtered through celite. The solution was cooled to 0°C, acidified with gaseous HCl, and the solvent
133
was evaporated under reduced pressure. The resulting oil was repeatedly triturated with diethyl ether to remove benzyl alcohol and to induce crystallisation of (141.HC1) as a hygroscopic pale yellow solid (54 mg, 78%) which was recrystallised from toluene.
ÔH (300 MHz, CDCI3): 1.61 (s, 3H), 1.76 (m, 6H), 1.95 - 2.29 (m, 3H), 2.30 (m, IH), 2.60 (m, 2H), 2.73 (d, n h = 4.4 Hz, 3H), 4.08 (bm, IH), 11.62 (NH); in addition, small signals at 6 3.85 and 6 11.06 corresponded to H6 and the NH protons of the minor stereoisomeric quaternary salt. An 8:1 ratio of equitorial:axial stereoisomers was calculated from the signal integrations.
5c (75 MHz, CDCI3), Major stereoisomer: 22.7 & 23.6 (2 x CH2), 24.8 (CH3), 28.1 & 28.2, (2 X CH2), 29.3 (CH3 ), 34.6 (CH2 ), 37.1 (CH2), 62.7 (CHN), 70.0 (C).
Vmax (CH2CI2): 3680W, 3380br, 3025m, 2940s, 2870m, 2390vbr, 1465m, 1445m, 1380m, 1265W, 1215w, 1205w, 1100m, 1085w, 1065w, 1045w, lOlOw, 995w, 970w, 905s, 845w cm'^
% (%): 153 (M+- HCl, 28), 152 (5), 126 (6), 124 (20), 111 (12), 110 (95), 98 (6 ), 97 (69), 96 (100), 82 (6 ), 81 (6 ), 71 (5), 56 (26), 55 (10).
C10H19N [M+ - HCl] requires '"/z 153.1517; observed: 153.1516.
1 -Methvl-9-azabicyclol4.2.1 Inonane (122)A solution of (136) (69 mg, 0.25 mmol) in dry methanol (6 ml) was hydrogenolysed with a catalytic amount of 5% palladium on charcoal at 1 atmosphere. After 4 hr the solution was filtered through a Millipore 0.2p Millex-FG disposable filter unit giving a colourless solution. This was acidified with gaseous hydrogen chloride before
evaporation of solvent under reduced pressure yielding (122.HC1) (42 mg, 95%) as a hygroscopic pale yellow solid. The 300 MHz ^H NMR spectrum was identical to that of a sample prepared by Smith.^^
ÔH (300 MHz, CDCI3): 1.46 - 2.35 (series of m, IIH [plus 6 1.71 (s, 3H)]), 2.51 (m, IH), 4.21 (m, IH, a-N), 9.23 (brm, IH, NHg), 9.66 (brm, IH, NHg).
134
4-( [B enzyloxycarbonvH aminolcyclooctanone-p-toluenesulphonvlhvdrazone ( 146)The ketone (133) (1.600 g, 5.82 mmol) was added to absolute ethanol (45 ml) and warmed until fully dissolved. p-Toluenesulphonylhydrazide (1.082 g, 5.82 mmol) was added with a catalytic quantity of p-toluenesulphonic acid. The solution was refluxed for 3 hr and the volume of solvent was then concentrated to two thirds of the original volume by distillation under reduced pressure. After refrigeration overnight, the crystals that formed were filtered and washed with cold ethanol and dried to afford (146) (1.688 g, 65%) as a crystalline white solid. M.p. 156 - 159°C.
ÔH (300 MHz, CD3SOCD3): 1.25 - 1.83 (series of m, 8H), 2.08 (m, IH), 2.03 - 2.42 (series of m, 3H [plus 2.73 (s, 3H)]), 3.46 (m, IH, a-N), 5.01 (s, 2H, CH^Ph), 7.29 - 7.42 (m, 5H plus part of AA’BB’, 2H), 7.75 (part of AA’BB’, 2H). The NH signals were not obseiwed.
V m a x (Nujol): 3360w, 3240w, 1690s, 1535w, 1335w, 1325w, 1310w, 1245w, 1160m, 1040w, 810w cm
""/z (%): 443 (M+, 1), 387 (9), 316 (13), 315 (17), 288 (43), 260 (26), 259 (100), 258( 100).
Found: C, 62.26; H, 6.74; N, 9.41%.C23H29N3O4S requires: C, 62.28; H, 6.59; N, 9.47%.
[(Benzyloxycarbonvl)aminolcvclooct-3-ene and -4-ene (147a,b)A solution of (146) (261 mg, 0.59 mmol) in THF (12ml) was cooled to -78°C and stiiTed under a nitrogen atmosphere. A solution of n-butyllithium (2.5M in hexane, 0.94 ml, 2.36 mmol) was injected and warmed to -10°C after 10 min. After a further 15 min the brick red solution was warmed to room temperature and quenched with water (200 pi). The solvent was concentrated under reduced pressure, the residual oil taken up in diethyl ether (15 ml) and washed with water ( 2x4 ml). The ethereal layer was separated, dried over anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure. The crude oil was purified by flash column chromatography using 1:5 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford a mixture of alkenes (147a,b) (64 mg, 42 %) as a crystalline white solid, which after recrystallisation from toluene and petroleum ether (b.p. 40 - 60°C), gave m.p. 59- 61
°C (mixture of 147a,b). The signals given in italics are common to both
135
regioisomers.
Sjj (300 MHz, CDCI3): 1.26 - 1.96 (series of m, 7H), 2.04 - 2.47 (series of m, 3H),
3.71 (brm, IH, a-N), 4.76 (brm, IH, HN), 5.07 (s, 2H, CHgPh), 5.57 - 5.80 (series of m,2H), 7.27-7.40 (m, 5H).
ÔC (75 MHz, CDCI3); 22.5, 23.3, 25.85, 25.9, 28.7, 31.0, 32.1, 34.7 & 35.5 (9 x CHg),51.2 (CHN), 51.6 (CHN), 66.5 (CHzPh), 126.0 (=€«2), 128.0, 128.1 & 128.5 (3 x aryl CH), 129.6,130.1 & 133.0 (3 x =CHz), 136.6 (aryl CHg), 155.3 (C=0).
Vmax (CH2CI2): 3420m, 3030m, 2940m, 2860w, 1720s, 1510s, 1450w, 1395w, 1355s, 1300W, 1225s, 1170s, 1125w, 1090w, 1025m, 970m, 935s, 900w, 880w, 855m, 820w
% (%): 259 (M+, 3), 231 (8), 168 (9), 146 (4), 124 (8), 92 (9), 91 (100).
C16H21NO2 [M+] requires % 259.1572; observed 259.1572.
Found: C, 74.15; H, 8.20; N, 5.39%.C16H21NO2 requires: C, 74.10; H, 8.16; N, 5.40%.
N-(Benzvloxvcaibonvl)-9-azabicyclor4.2. llnonane (148)Mercury(n) trifluroacetate (98 mg, 0.23 mmol) was added to a solution of (147a,b) (55 mg, 0.21 mmol) in acetonitrile (5 ml) and stirred for 18 hr, under a nitrogen atmosphere, as described for the preparation of (136). Following treatment with sodium borohydride (19 mg, 0.50 mmol) and purification by flash chromatography using 1:11 diethyl ether:petroleum ether (b.p. 40 - 60 °C) (148) was isolated (40 mg, 73 %) as a colourless oil. The figures in italics refer to signals common to both rotamers.
Ôjj (300 MHz, CDCI3): 1.34 - 1.71 (series of multiplets, 8H), 1.94 - 2.22 (series of m, 4H), 4.33 (m, 2H, a-N), 5.09 & 5.16 (ABq, J = 12.6 Hz, 2H, CH2Ph), 7.33 (m, 5H).
ÔC (75 MHz, CDCI3): 24.1 & 24.2 (2 x CH2), 29.3 & 29.7 (2 x CH2), 30.2 &31.3 (2 x CH2), 55.3 & 55.7 (2 x CHN), 127.65, 127.7 & 128.4 (3 x aryl CH), 137.2 (aryl C),755.7 (C=0).
136
Vmax (CH2CI2): 3030w, 2920s, 2860w, 1680s, 1415s, 1360m, 1330s, 1210w, 1080w, 1100s, 1030w, 990m, 945m, 905m, 870w cm '\
% (%): 259 (M+ 30), 172 (5), 168 (9), 159 (6), 158 (20), 124 (17), 92 (9), 91 (100).
C16H21NO2 [M+] requires ™/z 259.1572; observed 259.1573.
N-Methvl-9-azabicyclor4.2. llnonane (homotropane) (64)A solution of (148) (35 mg, 0.14 mmol) in dry diethyl ether (7 ml) and LAH (16 mg, 0.42 mmol) was refluxed under an nitrogen atmosphere for 45 min. The solution was cooled in an ice bath and excess hydride was destroyed by the dropwise addition of water-saturated diethyl ether. After complete decomposition the solution was dried
over anhydrous sodium sulphate and filtered through celite to afford a colourless solution. This was cooled to 0°C, acidified with gaseous hydrogen chloride and the solvent was evaporated under reduced pressure. The resultant oil was repeatedly triturated with a 1:1 mixture of diethyl ether:petroleum ether (b.p. 40 - 60°C) to remove benzyl alcohol and to induce solidification of (64.HC1) (23 mg, 97%) as a hygroscopic white solid. The hydrochloride salt (64.HC1) was dissolved in CDCI3, basified with gaseous ammonia and filtered to afford the free amine (64). The 300 MHz ^H NMR spectrum was identical to that of a sample prepared by Smith.^^
Ôfj (300 MHz, CDCI3): 1.37 - 1.63 (series of m, 8H), 1.79 - 1.86 (m, 2H), 2.12 - 2.24 (m, 2H), 2.42 (s, 3H), 3.25 (m, 2H, a-N).
% (%): 139 (M+ - HCl, 47), 110 (22), 97 (12), 96 (100), 83 (32), 82 (61).
9-Azabicvclo[4.2.llnonane (norhomotropane) (65)A solution of (148) (36 mg, 0.14 mmol) in absolute ethanol (5 ml) was hydrogenolysed in the presence of a catalytic quantity of 5% palladium on charcoal at 1 atmosphere. After 4 hr the solution was filtered through a Millipore 0.2p Millex-FG disposable filter unit giving a clear solution which was acidified with gaseous hydrogen chloride at 0°C. Distillation of the solvent under reduced pressure furnished
(6S.HC1) (20 mg, 89%) as a hygroscopic white solid, with an identical 300 MHz spectram to that of a sample prepared by Smith.^®
137
ÔH (300 MHz, CDCI3): 1.63 - 1.83 (series of m, 6H), 1.96 (m, 2H). 2.23 (m, 2H), 2.39 (m, 2H), 4.19 (m, 2H, a-N), 9.08 (brm, IH, NH), 9.95 (brm, IH, NH).
C/.y-4-r(BenzvIoxvcarbonvl)amino1cvclooct-2-enol (150)
To a solution of (103) (4.80 g, 18 mmol) in dry ethanol (41 ml) was added sodium phosphate (11.40 g, 80 mmol). The suspension was stirred under a nitrogen atmosphere for 5 min. Freshly prepared^^ and powdered 6% sodium amalgam (47 g) was added and the mixture was stirred for 1 hr. The solution was filtered through
celite and the solvent removed under vacuum. The residual oil was partitioned between water (30 ml) and dichloromethane (40 ml). The organic layer was separated and the aqueous layer was extracted further with dichloromethane (2 x 40 ml). The combined organic layers were dried over magnesium sulphate, filtered, and the solvent removed under vacuum. The resulting oil was triturated with petroleum ether (b.p. 40 - 60°C) to yield (150) (4.74 g, 98%) as a white solid (4.74 g, 98%) which was recrystallised from toluene and petroleum ether (b.p. 40 - 60°C) to afford a crystalline white solid, m.p. 127 - 128°C.
ÔH (300 MHz, CDCI3): 1.25 - 1.62 (m, 6H), 1.88 (m, 2H), 2.62 (brs, OH), 4.41 (brs, IH, a-N), 4.62 (brm, IH, a-OH), 5.01 (d, J = 5.7 Hz, NH), 5.06 (s, 2H), 5.25 (ddd, J =10.4, 8.3,1.5 Hz, IH), 5.60 (dd, J = 10.4,7.0 Hz, IH), 7.33 (m, 5H).
ÔC (75 MHz, CDCI3): 23.4, 24.0, 36.5 & 38.3 (4 x CHj), 49.4 (CHN), 66.6 (CHgPh),69.3 (CHO), 128.0 (2 x aryl CH), 128.4, (aryl CH), 129.4 & 139.4 (2 x =CH), 136.4 (aryl C), 155.7 (C=0).
Vmax (CH2CI2): 3600m, 3430m, 3420m, 2930s, 2860w, 1715s, 1505s, 1450w, 1395w, 1315brw, 1235m, 1210m, 1130w, 1085w, 1045m, 1015m, 950w, 860w cm \
% (%): 275 (M+, 1), 184 (19), 172 (6), 140 (13), 123 (5), 108 (9), 107 (5), 95 (4), 92(9), 91 (100), 80 (4), 79 (7).
Found: C, 69.87; H, 7.74; N, 5.08%.C16H21NO3 requires: C, 69.79; H, 7.69; N, 5.09%.
138
4-r(Benzvloxvcarbonvl)amino1cvclooct-2-enone (151) and N-benzvloxvcaibonvl-l-hvdroxv-9-azabicvclor4.2.nnon-7-ene (152)Barium manganate^^ (48.0 g, 180 mmol) was added to a stirred solution of (ISO) (4.56 g, 16.6 mmol) in dry dichloromethane (215 ml) under nitrogen. The suspension was stiiTed at room temperature for 48 hr. Dichloromethane (100 ml) was added with
stirring and the mixture was filtered through celite. The solvent was removed under vacuum and the resulting oil purified by flash chromatography using 1:4 diethyl ether:petrol (b.p. 40 - 60°C) to yield firstly (152) as a yellow oil. Further elution afforded (151) also as a yellow oil, the two tautomers having a combined weight of3.83 g, 85%. Leaving a solution of either tautomer to stand over a period of days resulted in equilibration to a 37:63 ratio of (151):(152) respectively (calculated from signal integrations). Data for the monocyclic (151) and bicyclic (152) tautomers aie given separately below; where signals due to two rotamers were visible, figures given italics are common to both rotamers.
Ôjj (300 MHz, CDCI3), Monocyclic tautomer: 1.44 - 2.01 (series of m, 6H), 2.54 (m, IH), 2.84 (m, IH), 4.95 (m, NH), 5.01 (m, IH, a-N), 5.12 (s, 2H), 6.09 (m, 2H, =CH).
Bicyclic tautomer: 1.27 -1.66 (series of m, 5H), 1.89 (m, IH), 2.02 (m, IH), 2.29 (m, IH), 4.71 (m, IH, a-N, major rotamer), 4.75 (m, IH, a-N, minor rotamer), 5.13, 5.20 (ABq, J = 12.2 Hz, 2H, CH^Ph), 5.71 (d, J = 6.2 Hz, IH, =CH, minor rotamer), 5.74
(dd, J = 6.2, 1.0 Hz, IH, =CH, major rotamer), 5.86 (dd, J = 6.2, 2.6 Hz, IH, =CH, major rotamer), 5.89 (dd, J = 6.2, 2.6 Hz, IH, =CH, minor rotamer), 7.30 - 7.37 (m, 5H).
(75 MHz, CDCI3), Monocyclic tautomer: 22.4, 22.6, 31.0 & 42.0 (4 x CH2), 50.6 (CHN), 67.0 (CH2Ph), 128.16, 128.22 & 128.5 (3 x aryl CH), 132.7 (=CH), 136.1 (aryl C), 144.1 (=CH), 155.6 (NC=0), 203.4 (a,p-unsaturated CO).
Bicyclic tautomer: 23.2, 23.6, 30.8 & 38.7 (4 x CH2), 60.8 (CHN), 66.4 (CH2Ph),95.7 (COH), 127.8, 128.0 & 128.4 (3 x aryl CH), 131.6 & 132.3 (2 x =CH), 136.4 (aryl C), 154.0 (C=0).
^max (CH2CI2), Monocyclic tautomer: 3430m, 2950m, 2860m, 1715s, 1660m, 1500m, 1350w, 1320brm, 1215m, 1170w cm \
139
Vmax (CH2CI2), Bicyclic tautomer: 3470brw, 2930m, 2850w, 1670s, 1415m, 1350m, 1320m, 1190m, 1125m, 1095m, 1040w, 1020w, 995m, 945w, 835w cm '\
"»/z (%): 273 (M+, 13), 229 (10), 186 (34), 165 (12), 138 (23), 137 (11), 124 (16), 122(22), 121 (21), 120 (17), 109 (26), 108 (87), 107 (61), 106 (10), 105 (30), 91 (100).
CjgHjgNOg [MT requires *”/z 273.1365; observed 273.136.
4-Methylene- F (benzvloxvcarbonyl)amino1cvclooct-2-ene (153)
The exo-methylene derivative (153) was prepared using the method described for (135) using sodium hydride (80% dispersion, 384 mg, 12.8 mmol) in dry DMSO (8 ml), methyltriphenylphosphonium bromide (4.75 g, 13 mmol) in DMSO (13 ml), and (151 ^ 152) (698 mg, 2.56 mmol) in DMSO (8 ml). After flash chromatography using 1:9 diethyl ether:petrol ether (b.p. 40 - 60°C) (153) was obtained as a white solid (397 mg, 57%) which had m.p. 94 - 95°C after reciystallisation from petroleum ether (b.p. 80 - 100°C).
ÔH (300 MHz, CDCI3): 1.31 (m, IH), 1.60 (m, 4H), 1.90 (m, IH), 2.36 (m, IH), 2.68 (m, IH), 4.87 (brs, IH, =CH2), 4.90 (s, NH), 4.96 (d, J = 1.0 Hz, IH, =CH2), 5.08, (s,
2H, CHzPh), 5.09 (m, IH, =CH), 5.15 (m, IH, a-N), 6.18 (d, J= 11.5 Hz, IH, =CH), 7.32 (m, 5H).
ÔC (75 MHz, CDCI3): 21.6, 28.1, 33.6 & 34.4 (4 x CHg), 49.6 (CHN), 66.6 (CHgPh),118.9 (C=CH2), 128.0, 128.1 & 128.4 (3 x aryl CH), 130.1 (=CH), 134.1 (=CH),136.5 (aryl C), 146.0 (CzzCH^), 155.7 (C=0).
V m a x (CH2CI2): 3430m, 3010, 2940m, 2850w, 1715s, 1590w, 1505s, 1470w, 1460w, 1320w, 1220m, 1040m, 1030m, 975w, 895w cm"\
""/z (%, Cl): 272 (MH+, 100), 228 (91), 211 (20), 136 (22), 121 (36), 108 (25), 91
(17).
C17H22NO2 [MH+] requires ”*/z 272.1651; observed 272.165.
Found: C, 75.19; H, 7.97; N, 5.17%.C17H21NO2 requires C, 75.24; H, 7.80; N, 5.16%.
140
N-CBenzyloxycarbonyl)-1 -methyl-7B-hvdroxv-9-azabicyclor4.2.11nonan-7-ol (154a)
and N-(benzyloxycarbonvl)-l-methyI-8B-hvdroxv-9-azabicyclor4.2.11nonan-7-ol (154b)The cyclisation of (153) was carried out according to the method described for preparing (136) using mercury(II) trifluoroacetate (216 mg, 0.51 mmol) and (153) (131 mg, 0.48 mmol) in dry acetonitrile (8 ml). The reduction was carried out using sodium borohydride (37 mg, 0.98 mmol) in dry THF (14 ml) to give (154a,b) (76 mg, 54%) after flash chromatography using 2:3 diethyl ether:petrol ether (b.p. 40 - 60°C).
ÔH (300 MHz, CDCI3): 1.21 - 1.66 (series of m, including CH3, 8H), 1.79 - 2.41 (series of m, 5H), 2.58 (s, IH, exch), 3.86 - 4.40 (series of m, a-N & a-OH, 2H), 5.08 (m, 2H, CH2Ph), 7.33 (m, 5H). The high-resolution ^H and NMR spectra of the mixture were complicated at ambient temperature due to slow rotation around the N-CO bond.
ôjj (300 MHz, 363K, Dg-toluene): 1.10 - 1.81 (series of m, including CH3, 13H), 5.02 & 5.12 (ABq, J = 12.3 Hz, 2H), 6.98 - 7.27 (m, 5H).7-Hydroxy (154a): 2.06 (dd, J = 14.4, 6.7 Hz, IH, p-OH), 3.64 (brd, J = 6.7 Hz, IH,
a-OH), 4.14 (m, IH, a-N).Double irradiation at 6 4.14 caused the brd at 6 3.64 to simplify to a dd (J = 6.7, 1.9 Hz) indicating a small vicinal coupling between a-OH and a-N protons (J < 1 Hz). This confirmed the C7 placing of the hydroxyl group. Double irradiation at 5 3.63 removed the vicinal coupling of 6.7 Hz at Ô 2.06 leaving a doublet (J = 14.4 Hz).8-Hydroxy (154b): 3.74 (dd, J = 6.8,4.8 Hz, IH, a-OH), 4.32 (m, IH, a-N).
Double iiTadiation at 8 4.32 resulted in no pertubation at 8 3.74 confirming the Cgplacing of the hydroxy group.
Vmax (CH2CI2): 3440br, 2930s, 2960m, 1680s, 1575w, 1530w, 1500w, 1450m, 1335m, 1235m, 1200m, 1160m, 1135m, 1050m cm \
% (%): 289 (M+, 17), 271 (6), 228 (6), 184 (8), 154 (24), 110 (11), 108 (13), 107(12), 106 (13), 105 (12), 91 (100).
CJ7H23NO3 [M+] requires ™/z 289.1680; observed 289.168.
141
N-(Benzvloxvcarbonvl)-l-methvl-9-azabicvclor4.2.11non-7-ene (155)The cyclisation of (153) was carried out according to the method described for preparing (136) using mercury(II) acetate (190 mg, 0.60 mmol) and (153) (80 mg,
0.30 mmol) in dry acetonitrile (19 ml). Reduction with sodium borohydride (23 mg, 0.61 mmol) in dry THF (12 ml) gave (155) (36 mg, 45%) as an oil after flash
chromatography using 1:9 diethyl ether:petroleum ether (b.p. 40 - 60°C) together with unchanged (153) (26 mg). Chemical shifts in italics refer to overlapping signals due to major and minor rotamers.
Ôjj (300 MHz, CDCI3): 1.26 -1.61 (m, 6H), 1.52 (s, 3H, minor rotamer), 1.65 (s, 3H, major rotamer), 1.97 (m, IH), 2.16 (m, IH), 4.72 (ddd, J = 6.0, 2.6, 1 Hz, IH, major rotamer), 4.79 (brd, J » 5.5 Hz, IH, minor rotamer), 5.07, 5.18 (ABq, J = 12.5 Hz, 2H, CH^Ph, major rotamer), 5.18 (s, 2H, CH2Ph, minor rotamer), 5.47 (d, J = 6.2 Hz, IH, minor rotamer), 5.49 (dd J = 6.2, 0.8 Hz, IH, major rotamer), 5.64 (dd, J = 6.2, 2.6 Hz, major rotamer), 5.67 (dd, J » 6.2, 2.6 Hz, IH, minor rotamer), 7.27 - 7.37 (m, 5H).
ÔC (75 MHz, CDCI3), Major rotamer: 23.7 (CH2), 24.3 (CH3), 24.8, 31.7 & 37.0 (3 x CH2), 62.8 (CHN), 66.1 (CH2Ph), 68.8 (CCH3), 127.75 (=CH), 127.8, 128.0 & 128.4 (3 X aryl CH), 136.4 (=CH), 137.0 (aryl C).
Minor rotamer: 23.4 & 25.0 (2 x CH2), 25.6 (CH3), 30.3 & 38.1 (2 x CH2), 63.8 (CH), 66.6 (CHgPh), 68.0 (CCH3), 127.7 (aryl CH), 127.8 (=CH), 128.1 & 128.5 (2 x aryl CH), 137.1 (=CH), 137.1 (aryl C); the CO signals were too weak to be resolved.
Vmax (CH2CI2): 3025m, 2865w, 1695s, 1630w, 1570w, 1560w, 1550w, 1520w, 1510w, 1505w, 1445w, 1405s, 1235m, 1155w, 1100m, 1040m, 935 cm"\
% (%): 271 (M+, 5), 185 (7), 92 (8), 91 (42), 44 (39), 32 (100).
CJ7H21NO2 [M^ requires "'/z 271.1572; observed 271.157.
N-(Benzvloxvcarbonvl)-l-methyl-9-azabicvclor4.2.11nonan-7-one (156a) and -8-one (156b)
A solution of (154) (33 mg, 0.11 mmol) was oxidised with Jones reagent^^ using the
142
procedure described for the conversion of (132) into (133). The regioisomeric mixture of ketones (156a,b) (32 mg, 98%) was obtained as a pale yellow oil which
was pure enough for spectroscopic analysis without chromatography. Signals common to both isomers are quoted in italics.
Ôjj (90 MHz, CDCI3): 0.70 - 2.80 (series of m including CH3 signals, 13H), 4.10 - 4.70 (IH, a-N), 5.00 - 5.25 (m, 2H, CHzPh), 7.30 (brs, 2H).
Ôjj (300 MHz, 363K, Dg-toluene): 0.75 - 2.23 (series of m, lOH, [including Ô 1.56 (s,
3H)]), 4.86 & 5.17 (ABq, J = 12.0 Hz, 2H, CHjPh), 6.98 - 7.24 (m, 5H).7-Keto (156a): 1.83 & 2.02 (ABq, J = 18.0 Hz, 2H, a-C=0), 4.02 (d, J = 8.4 Hz, IH, a-N).Double irradiation at Ô 4.02 sharpened one half of the ABq (w-coupling) but the lack of vicinal coupling confirmed the position of the C=0 at C7.8-Keto (156b): 2.18 (dd, J = 17.4, 8.7 Hz, IH, a-C=0), 4.32 (vbrm, IH, a-N).
Double irradiation at Ô 4.32 removed the vicinal coupling of 8.7 Hz at ô 2.18, leaving a d (J = 17.4 Hz), confirming the placing of the CH2 at C7 and the C=0 at Cg.
Vmax (CH2CI2): 2930m, 2860W, 1755s, 1695s, 1445w, 1395m, 1355w, 1335m, 1325w, 1260brw, 1205w, 1170w, 1110m, 1050m, 1025w, 965w, 950w, 930w, 905w cm'^
(%): 287 (M+, 4), 259 (7), 151 (6), 149 (4), 124 (11), 92 (9), 91 (100), 65 (6).
CJ7H21NO3 [M+] requires ™/z 287.1521; observed 287.1521.
N-(Benzvloxvcarbonvl)-l-methyl-73,83-epoxv-9-azabicvclor4.2. llnonane (165) MCPBA (50 - 60 % purity, 49 mg, 0.16 mmol) was added to a stirred solution of (155) (41 mg, 0.14 mmol) in dichloromethane (6 ml). After stirring for 16 hr a further portion of MCPBA (10 mg, 0.03 mmol) was added and stirred for a further 24 hr. The solution was evaporated under reduced pressure and the residual oil was taken up in diethyl ether (14 ml), washed with saturated sodium bicarbonate solution ( 4 x 3 ml)
and water (3 ml). The ethereal layer was dried over anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure. The resultant oil was purified by flash chromatography using 1:9 diethyl ether:petroleUm ether (b.p. 40 - 60°C) to yield (165) (41 mg, 94 %) as a colourless oil. Chemical shifts in italics refer to overlapping signals due to major and minor rotamers.
143
Ôjj (300 MHz, CDCI3): 1.26 -1.65 (series of m, 6H), 1.54 (s, 3H, minor rotamer), 1.67 (s, 3H, major rotamer), 1.86 - 2.27 (series of m, 2H), 3.22 (d, J = 3.1 Hz, IH, HCO, minor rotamer), 3.23 (dd, J = 3.1, 0.4 Hz, IH, HCO, major rotamer), 3.32 (d, J = 3.1 Hz, IH, HCO, major rotamer), 3.33 (d, J = 3.1 Hz, IH, HCO, minor rotamer), 4.36 (d, J = 6.8 Hz, a-N, major rotamer), 4.42 (d, J = 6.5 Hz, a-N, minor rotamer), 5.00 & 5.16 (ABq, J = 12.3 Hz, 2H, CHzPh), 7.27 - 7.40 (m, 5H).
ÔC (75 MHz, CDCI3), Major rotamer: 21.2 (CH3), 23.66, 24.5, 29.6 & 35.2 (4 x CHj),56.5 (CHO), 57.6 (CHO), 62.3 (CHN), 63.2 (CN), 66.4 (CHgPh), 127.9 (2 x aryl CH),128.3 (aryl CH), 735.7 (aryl C), 155.0 (C=0).
Minor rotamer: 22.5 (CH3), 23.75, 24.6, 28.3 & 36.5 (4 x CHg), 56.1 & 58.3 (2 x CHO), 62.3 (CHN), 63.3 (CN), 66.9 (CH^Ph), 127.9 (2 x aryl CH), 128.3 (aryl CH), 735.7 (aryl C), 155.0 (C=0).
Vmax (CH2CI2): 3070W, 3045w, 3015w, 2920s, 2850m, 1710brs, 1580w, 1570w, 1535w, 1495w, 1450m, 1425m, 1395m, 1360m, 1350m, 1275w, 1260w, 1230w,
1210m, 1180m, 1145m, 1125m, 1070m, 1030w, lOOOw, 960w, 945w, 925w cm"\
^/z (%): 287 (M+, 30), 180 (37), 153 (13), 152 (41), 124 (15), 110 (21), 108 (12), 107 (11), 106(13), 105 (10), 91 (100).
CJ7H21NO3 [M+] requires ^/z 287.1521; observed 287.1521.
N-Benzvloxvcarbonvl-l-methoxvethvoxvmethoxv-9-azabicvclor4.2.l1non-7-ene(185)A solution of n-butyllithium (2.5 M in hexane, 1.44 ml, 3.60 mmol) was injected into a stined solution of (151 ^ 152) (893 mg, 3.27 mmol) in THF (25 ml) at 0°C under a nitrogen atmosphere. After 15 min, MEM-Cl (0.52 ml, 4.58 mmol) was injected. The solution was allowed to warm to room temperature, then heated at reflux for 5 hr. The solvent was removed under reduced pressure and the residue dissolved in dichloromethane (55 ml), washed with water (3 x 20 ml), separated and dried over anhydrous magnesium sulphate. Filtration and evaporation of solvent afforded a yellow oil which was purified by flash chromatography using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) yielding (185) (863 mg, 73%) as an oil.
144
Ôjj (300 MHz, CDCI3), Slow rotation about the N-CO bond was observed and signals due to the two rotamers are shown separately where they were resolved; where signals overlapped, 8 values are shown in italics; 1.29 - 1.60 (series of m, 5H), 1.81 (m, IH),2.03 (m, IH), 2.29 (m, IH), 2.50 (m, IH), 3.35 (s, 3H, MeO), 3.38 (s, 3H, MeO), 3.53 (m, 2H), 3.63 - 3.87 (series of m, 2H), 4.58 and 4.88 (ABq, J = 6.9 Hz, 2H,
O-CH2-O), 4.70 (m, IH, a-N), 4.72 and 4.91 (ABq, J = 7.2 Hz, 2H, O-CHg-O), 5.15 (ABq, J = 12.4 Hz, 2H, benzyl CH2), 5.21 (ABq, J = 12.8 Hz, 2H, CH2Ph), 5.77 (dd, J = 6.2, 0.6 Hz, IH), 5.85 (dd, J = 6.2, 2.6 Hz, IH), 5.88 (dd, J = 6.2, 2.6 Hz, IH), 7.34 (m, 5H).
ÔC (75 MHz, CDCI3): 22.7 & 23.0 (2 x CHg), 23.1 & 23.4 (2 x CH2), 30.1 & 31.5 (2 x CH2), 35.7 & 36.9 (2 x CH2) 58.8 & 58.9 (2 x CH2), 61.0 & 61.6 (2 x CHg), 66.2 &66.6 (2 X CH2), 67.4 & 67.5 (2 x CH2), 71.7 (CH2), 89.9 & 90.0 (2 x CHg), 98.1 & 99.0 (2 X COH), 127.8 (aryl CH), 127.9 & 128.1 (2 x aryl CH), 128.4 (aryl CH), 131.0 & 131.4 (2 X =CH), 133.1,133.8 (2 x =CH), 136.7{C), 152.9 & 154.3 (2 x C=0).
Vmax (thin film): 3080w, 3060w, 3020w, 2920s, 2870s, 2800w, 1695brs, 1620w, 1580W, 1490w, 1440m, 1395s, 1345s, 1300s, 1250m, 1195m, 1170m, 1090brs, 1020s,
980s, 935m, 845w, 830w, 805w, 790m, 770m, 730m, 695m cm't.
»"/z (%): 361 (Mt, 32), 285 (26), 274 (35), 273 (39), 256 (67), 241 (50), 228 (29), 212 (34), 199 (42), 186 (34), 165 (24), 150 (100), 138 (42).
C20H27NO5 [M+] requires ™/z: 361.1889; observed 361.189.
N-Methvl-9-azabicyclol4.2.11non-7-en-l-ol (186 ^ 187)In flame-dried apparatus, a solution of (185) in THF (12 ml) was injected into a slurry of LAH (123 mg, 3.24 mmol) under nitrogen at 0°C. The mixture was subsequently
heated at reflux for 4 hr. Quenching of excess hydride with water-saturated ether, drying over sodium sulphate, filtration through celite and distillation of solvent afforded an oil (186 ^ 187) (292 mg) which was deprotected without further purification. The oil was dissolved in dichloromethane (11 ml) and trifluroacetic acid (0.52 ml, 6.77 mmol) was added. After stirring for 2 hr, water (15 p,l) was added, the mixture was stirred for a futher 10 min, and the solvent was removed under vacuum ensuring complete removal of methanal. The residue was dissolved in water (3 ml)
145
and extracted with diethyl ether ( 2 x 4 ml) to remove benzyl alcohol. The pH of the aqueous layer was adjusted to pH 9 by addition of sodium hydroxide solution (2 M) and the product was extracted into dichloromethane ( 5 x 4 ml). Futher basification (pH 11) and extraction afforded a yellow solid (total 95 mg) after drying with MgSO^
and evaporation of solvent. The crude product was recrystallised from petrol (b.p. 80 - 1G0°C) to afford (186 ^ 187) (74 mg, 45% overall) as an off-white solid, m.p. 94 - 110°C.
5h (300 MHz, 298 K, CDCI3); 1.59 (m, 6H), 1.86 (m, IH), 2.17 (m, IH), 2.55 (s, 3H, CH3), 3.87 (m, IH, a-N), 5.90 (d, J = 7.8 Hz, IH), 6.01 (dd, J = 7.8, 3.4 Hz, IH).
Ôq (75 MHz, 298 K, CDCI3), The italicised signals were broadened at this temperature); 22.8 (CH^), 23.3 (CHj), 29.7 (CHg), (CH2), 37.9 (CH2), 62.4 (CH),136.3 (CH).
ÔH (300 MHz, 223 K CDCI3): 1.45 - 2.18 (series of m, 8 H), 2.55 (s, 3 H), 3.78 (m, 1 H), 5.86 (hr d, J « 5.8 Hz), 5.98 (vbr d, J « 5.8 Hz).
ÔC (75 MHz, 223 K CDCI3): 22.5, 23.6, 28.3, 28.7 & 35.6 (5 x CH2), 62.9 (CH), 94.8 (COH), 132.8,137.4 (2 x =CH).
Vmax (CH2CI2): 3570w, 3055vbrs, 2940s, 1705m, 1645s, 1610s, 1400m, 1360w, 1305m, 1215w, 1135m, 1090m, 1055m, 1020m, 910m cm'^
™/z (%): 153 (M+, 18), 125 (11), 110 (100), 97 (38), 96 (39), 70 (17), 68 (14).
C9H15NO [M+j requires ™/z 153.1154; observed 153.1156.
N-Benzvloxvcarbonvl-l-methoxvethoxvmethoxv-9-azabicyclo[4.2.l1nonane (188) n-Butyllithium (2.5 M in hexane, 0.58 ml, 1.45 mmol) was injected into a stirred solution of (133) (364 mg, 1.32 mmol) in THF (13 ml) at 0°C. After 15 min, MEM-Cl (0.21 ml, 1.83 mmol) was added and the solution was heated under reflux for 3 hr. The bulk of the solvent was distilled in vacuo and the residual oil dissolved in dichloromethane (22 ml) and washed with water (2 x 10 ml). The organic layer was separated, dried with anhydrous magnesium sulphate, filtered and the solvent distilled. The crude product was purified by flash chromatography using 2:3 diethyl
146
ether:petroleum ether (b.p. 40 - 60°C) to afford (188) (336 mg, 70%) as a pale yellow oil.
Ôjj (300 MHz, CDCI3), Broadening and/or signal overlap due to slow N-CO rotation is indicated using italics: 1.24 - 2.63 (series of m, 12H), 3.33 (s, 3H), 3.36 (s, 3H), 3.51 (m, 2H), 3.68 (m, 2H), 3.81 (m, 2H), 4.33 (m, IH, a-N), 4.58, 4.77 & 4.91 (series of m, 2H, O-CH2-O & a-N, IH), 5.15 (m, 2H, CH2Ph), 7.32 (m, 5H).
ÔC (75 MHz, CDCI3): 22.9 (CH2), 23.6 (CH2), 27.1 & 27.7 (2 x CH2), 33.3 & 35.0 (2 X CH2), 36.0 & 36.6 (CH2) 37.4 & 38.4 (2 x CH2), 56.2 & 57.0 (2 x CH), 58.8 {CH3 ),66.2 & 66.6 (2 x CH2Ph), 66.7 & 67.4 (CH2), 71.7 (CH2 ), 90.0 (CH), 95.6 & 96.3 (C),127.8,127.9 & 128.3 (3 x aryl CH), 136.8 (aryl C), 153.6 & 154.9 (2 x C=0).
Vmax (CH2CI2): 2930m, 2895w, 1695s, 1500w, 1450w, 1395m, 1355w, 1335w, 1315w, 1275brw, 1215w, 1160w, 1115m, 1095m, 1030s, 995w, 955w, 935w, 910w,
850w cm’ .
% (%): 363 (M+, 2), 274 (14), 258 (25), 214 (10), 168 (18), 152 (39), 140 (11), 124(10), 91 (100).
C20H29NO5 [M+] requires 363.2050; observed 363.2049.
N-Methvl-9-azabicvclol4.2.1 Inonan-1 -ol (189 ^ 190)The protected carbamate (188) (290 mg, 1.20 mmol) was reduced with hydride and deprotected with trifluoroacetic acid using identical procedures to those described for the preparation of (186 ^ 187). The product (189 ^ 190) was isolated as an impure oil (21 mg, 17%). An improved route to (189 ^ 190) from (192) is deseribed below.
9-Azabicvclol4.2. llnonan-1 -ol (176 ^ 177)
A solution of (133) (330 mg, 1.20 mmol) in ethanol (18 ml) was hydrogenolysed with a catalytic amount of 5% palladium on charcoal at 1 atmosphere. After 2 hr the solution was filtered through a Millipore 0.2p Millex-FG disposable filter unit and the solvent distilled under reduced pressure to afford (176 ^ 177) (165 mg, 98%) as an off white solid. A sample prepared using this method had identical spectroscopic properties to data recorded by Smith.^^
147
Sh (90 MHz, CDCI3): 1.55 (m, 6H), 2.10 (m, 6H), 2.80 (brs, 2H, exch, OH & NH), 3.25 (brm, IH, a-N).
Cf5-4-(Methvlamino)cvclooctanol ( 192)
A 100 ml 2-necked flame dried flask was charged with LAH and stirred with dry THF (2 ml). The system was alternately charged and evacuated with nitrogen and cooled to 0°C. A solution of (132) (1.353 g, 4.88 mmol) in THF (24 ml) was injected. After complete addition, the mixture was refluxed for 1 hr. Excess hydride was destroyed by dropwise addition of water-saturated diethyl ether, the solution dried with sodium sulphate and filtered through celite. The solvent was distilled under reduced pressure and the residual oil purified by flash chromatograpy using 4:1 dichloromethane : methanol (saturated with ammonia) to afford product (192) (709 mg, 92%) as a colourless oil which was identical to a sample prepared previously by a different route. ^
ÔH (90MHz, CDCI3): 1.20 - 1.85 (m, 12H), 2.25 (s, 3H), 2.40 (m, IH, a-N), 3.30 (brs, exch, 2H), 3.70 (m, IH, a-0).
N-Methvl-9-azabicvclor4.2.llnonan-1 -ol (189 ^ 190)
To a stirred solution of (192) (279 mg, 1.80 mmol) in propanone (12 ml) was oxidised by the dropwise addition of Jones reagent.® After 25 min, isopropanol was added until a permanent green colouration remained and the solution was then filtered through celite. The inorganic residues were washed further with methanol (2 x 10 ml). The solvent was distilled in vacuo and the residue dissolved in water (7 ml) and basified to pH 12 using sodium hydroxide solution (2 M). The water was removed in vacuo and the inorganic solids extracted with chloroform (80 ml), using a Soxhlet apparatus, for 18 hr. Distillation of solvent afforded (189 ^ 190) as a waxy yellow solid (189 mg, 69%).
ÔH (300 MHz, 298 K, CDCI3,): 1.46 (m, IH), 1.61 (m, 6H), 1.86 (m, IH), 2.11 (m, 5H), 2.45 (s, 3H), 3.17 (m, IH), 4.30 (brs, IH exch).
ÔC (75 MHz, 298 K, CDCI3), Broad signals are italicised: 23.6 (CHg), 24.0 (CH2),26.8 (CH2), 30.1 (CH3), 32.0 (CH2), 38.4 (CH2), 39.8 (CH2), 58.7 (CH).
Sh (300 MHz, 243 K, CDCI3): 1.49 (m, 2H), 1.63 (m, 5H), 1.94 (m, 4H), 2.20 (m.
148
2H), 2.49 (s, 3H), 3.34 (m, IH), 6.18 (vbrs, IH).
ÔC (75 MHz, 243 K, CDCI3): 22.4, 23.4 & 25.9 (3 x CHg), 28.8 (CH3), 31.9, 37.4 &39.4 (3 X CH2), 58.1 (CHN), 92.0 (COH).
Vmax (CH 2 CI2 ): 3180vbr, 2940s, 2870m, 2800w , 1695m, 1470w, 1450w, 1365w,
1340w, 1230W, 1210W, 1180w, 1155w, 1130w, 1080w, 1055w, 1045w, 1005w, 975w ,
940w , 915w , 865w , 835w , 815w c m '\
(%): 155 (M+, 8), 149 (13), 141 (6), 137 (3), 126 (17), 112 (20), 108 (13), 98 (32), 94(18), 70(100).
C9H17NO [M+] requires ™/z 155.1310; observed 155.1311.
Ci.y-4-(Methvlamino)cvcIooct-2-enol (193)The protected amino-alcohol (150) (308 mg, 1.12 mmol) in dry THF (6 ml) was added to a slurry of LAH (83 mg, 2.18 mmol) in dry THF at 0°C following thorough evacuation/purging of the reaction flask with nitrogen. The mixture was heated at reflux for 2.5 hr and excess hydride was then destroyed using water-saturated diethyl ether. Filtration through celite and distillation of solvent afforded a white solid which
was repeatedly triturated with a 1:1 mixture of diethyl ether:petrol (b.p. 40 - 60°C) to remove benzyl alcohol. The product (193) was obtained as a white solid, (145 mg, 84%), m.p. 123 - 125°C showing identical spectroscopic properties to a sample prepared previously by a different route^^ (lit. m.p. 125 - 126°C).
Ôjj (90 MHz, CDCI3): 1.20 - 1.95 (series of m, 8H), 2.20 (brs, exch, 2H), 3.40 (m, IH, a-N), 4.55 (m, IH, a-O), 5.40 (t, J= 11 Hz, IH), 5.70 (dd, J = 11,7 Hz, IH).
N-Methyl-9-azabicvclol4.2.11non-7-en-l-ol (186 ^ 187)
The allylic alcohol (193) (297 mg, 1.92 mmol) was treated with Jones reagent® using an identical procedure to that described below for conyersion of (192) into (189 ^
190). Extraction into chloroform using a Soxhlet apparatus gaye a yellow solid which was reciystallised from toluene and petroleum ether (b.p. 60 - 80°C) to afford (186 187) (176 mg, 63%) as a slightly yellow solid which was identified by comparison of spectra with those of the sample obtained earlier from (185).
149
N-Benzvloxvcarbonyl-7|3.8S-epoxv-9-azabicvclor4.2.llnonan-l-ol (194)MCPBA (50-60% purity, 2.680g, 8.57 mmol) was added in portions to a stirred solution of (151 ?=* 152) (1.942 g, 7.11 mmol) in dichloromethane (67 ml). After stirring for 24 hr a further portion of MCPBA (0.80 g, 2.54 mmol) was added and
stined for a further 48 hr. The solyent was concentrated under reduced pressure and the residual oil was dissolyed in diethyl ether (65 ml) and repeatedly washed with saturated sodium bicarbonate solution (5 x 20 ml) and brine (20 ml). The ethereal layer was separated, dried oyer anhydrous magnesium sulphate, filtered and the
solvent distilled under reduced pressure. After flash column chromatography using 3:2 diethyl ether:petroleum ether (b.p. 40 - 60°C) (194) was isolated as a white solid
(1.626 g, 79%) which had m.p. 66 - 68°C after recrystallisation from toluene and petroleum ether (b.p. 40 - 60°C).
Ôjj (300 MHz, CDCI3), Signals conunon to both rotamers are quoted in italics: 1.19 - 1.62 (series of m, 5H), 2.04 (m, 2H), 2.22 (m, IH), 3.33 (d, J = 3.2 Hz, IH, HCO, major rotamer), 3.37 (d, J = 3.2 Hz, IH, HCO, minor rotamer), 3.46 (d, J = 3.2 Hz, IH, HCO, minor rotamer), 3.48 (d, J = 3.2 Hz, IH, HCO, major rotamer) 4.35 (d, J =6.9 Hz, IH, a-N major rotamer), 4.41 (d, J = 6.2 Hz, IH, a-N minor rotamer), 5.05 & 5.18 (ABq, J = 12.3 Hz, 2H, CH2Ph, major rotamer), 5.13 & 5.23 (ABq, J = 12.3 Hz, 2H, CH2Ph, minor rotamer), 7.33 (m, 5H).
6c (75MHz, CDCI3), The heavy weighting towards the major rotamer resulted in the minor rotamer signals being too weak to be observed): 23.2 (2 x CH2), 29.1 & 37.6 (2
X CH2), 54.1 & 55.7 (2 X CHO), 58.0 (CHN), 66.8 (CH2Ph), 91.4 (COH), 127.9, 128.1 & 128.5 (3 X aryl CH), 136.0 (aryl C), 155.9 (CO).
Vmax (CH2C1)2: 3450brm, 3090w, 3050w, 3020w, 2920s, 2850m, 1670s, 1490w, 1420s, 1390s, 1345s, 1255w, 1235w, 1220w, 1190m, 1145m, 1110m, 1085m, 1050m, 1020m, 990m, 970m cm"\
% (%): 289 (M+, 27), 271 (19), 245 (17), 227 (27), 155 (25), 154 (76), 126 (49), 113 (19), 112(59), 108 (61), 91 (100).
CjgHjgNO^ [M+] requires 289.1314; observed 289.131.
150
Found: C, 66.46; H, 6.58; N, 4.66%.CjgHjgNO^ requires: C, 66.42; H, 6.62; N, 4.84%.
N-Methvl-73,83-epoxy-9-azabicvclor4.2.llnonan-l-ol (196)A solution of (194) (213 mg, 0.74 mmol) in dry THF (4 ml) was injected into a stined slurry of LAH in dry THF (2 ml). The solution was refiuxed for 30 min under a nitrogen atmosphere and excess hydride was quenched by the addition of water-saturated diethyl ether and then dried with anhydrous sodium sulphate. The colourless solution was filtered through celite and the solyent was evaporated under
reduced pressure to afford an oil. Repeated trituration with petroleum ether (b.p. 40 - 60°C) to remove benzyl alcohol gave (196) (89 mg, 71%) as a crystalline white solid having m.p. 119 - 120 °C (decomp.) after recrystallisation from ethanol.
5h (300 MHz, CDCI3): 1.55 (m, 5H), 1.97 (m, 3H), 2.52 (s, 3H), 2.99 (brs, exc, IH),3.34 (d, J = 3.3 Hz, IH, HCO), 3.38 (d, J = 3.3 Hz, IH, HCO), 3.41 (d, J = 4.9 Hz, IH, a-N).
8c (75MHz, CDCI3): 22.7 & 24.0 (2 x CHj), 28.9 (CH3), 29.1 & 35.0 (2 x CHg), 55.3 (CHN), 57.5 & 59.9 (2 x CHO), 89.6 (COH).
Vmax (CH2C1)2: 3560w, 2930s, 2870m, 1465w, 1440w, 1370w, 1340w, 1240w, 1210w, 1190w, 1105w, 1080m, 1020m, 945w, 880m, 860w, 850w, 830w cm"\
™/z (%): 169 (M+, 72), 149 (17), 140 (46), 127 (26), 126 (78), 113 (36), 112 (100), 98 (21), 84 (42), 70 (40).
Found: C, 63.82; H, 9.14; N, 8.31%.
C9H15NO2 requires: C, 63.88; H, 8.93; N, 8.23%.
73,83-Epoxv-azabicvclol4.2.11nonan-l-ol (197)A solution of (194) (981 mg, 3.39 mmol) in absolute ethanol (54 ml) was hydrogenolysed at 1 atmosphere with a catalytic quantity of 5% palladium on charcoal. After 2.5 hr the solution was filtered through a Millipore 0.2p Millex-FG disposable filter unit. Evaporation of ethanol under reduced pressure yielded (197) (510 mg, 97%) as a gum.
151
ÔH (300 MHz, CDCI3): 1.35 (m, IH), 1.47 - 1.73 (series of m, 4H), 1.82 (m, 2H), 2.00 (ddd, J = 13.9, 6.9,4.4 Hz, IH), 3.36 (m, 2H, HCO), 3.53 (d, J = 6.9 Hz, IH, a-N).
ÔH (300 MHz, CD3COCD3): 1.29 (m, IH), 1.46 (m, 2H), 1.58 - 1.79 (series of m, 4H), 1.89 (m, IH), 3.27 (d, J = 2.7 Hz, IH, HCO), 3.31 (d, J = 2.7 Hz, IH, HCO), 3.37 (d, J = 7.1 Hz, IH, a-N), 3.72 (brs, 2H, exc, OH & NH).
ÔC (75 MHz, CDCI3): 23.2, 23.7, 30.7 & 38.9 (4 x CHg), 54.6 (CHN), 58.4 & 59.4 (2 X CHO), 91.2 (COH).
Vmax (CH2C1)2: 3560w, 3300brs, 2930s, 2860s, 1425brm, 1340m, 1300m, 1265m, 1185m, 1105s, 1025m, 990m, 980m, 945m, 930m cm'^
""/z (%): 155 (M+, 47), 126 (100), 112 (74), 98 (72), 85 (28), 72 (58).
CgHj3N02 [M+] requires ™/z 155.0946; observed 155.0946.
N-(Benzvloxvcarbonvl)-6-aza-7-oxabicvclo[3.2.21non-8-ene (221) Tétraméthylammonium periodate (14.56 g, 0.055 mol) and cyclohepta-1,3-diene (43) (4.31 g, 0.046 mol) in dichloromethane (70 ml) was stirred at 0°C. A solution of benzyl-N-hydroxycarbamate (9.18 g, 0.055 mol) in dichloromethane (18 ml) was dripped in over 15 min. The mixture was warmed to room temperature on complete addition and stirred for a further 4 hr. The solution was filtered, washed with
saturated sodium thiosulphate solution (2 x 30 ml) and water (30 ml). The organic layer was separated, dried over anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure. The residual yellow oil was purified by flash chromatography using 2:8 diethyl ether:petroleum ether (b.p 40 - 60°C) to afford(221) (11.51 g, 97%) as an oily yellow solid. This compound was prepared earlier by Bathgate^^ in 90% yield.
ÔH (90 MHz, CDCI3): 1.45 (m, 2H), 1.80 (m, 4H), 4.75 (brm, 2H, a-O & a-N), 5.20 (s, 2H, CH2Ph), 6.20 (m, 2H), 7.30 (s, 5H).
N-(Benzvloxvcarbonvl)-6-aza-7-oxabicvclo[3.2.21nonane (222)To a stirred solution of potassium azodicarboxylate (12.74 g, 65.7 mmol) and (221) (3.40 g, 13.12 mmol) in methanol (90 ml) at 0°C was added glacial acetic acid (7.60
152
ml, 139.2 mmol) over 10 min. The mixture was wanned to room temperature and stirred for a further 2 hr. The reaction was quenched with water (3 ml), filtered and the bulk of the solvent distilled under reduced pressure. The residue was partitioned
between dichloromethane (70 ml) and saturated sodium bicarbonate solution (20 ml), washed with further bicarbonate solution (10 ml) and water (20 ml). The organic layer was separated, dried over anhydrous magnesium sulphate, filtered and the solvent removed under reduced pressure. The oil was purified by flash chromatography using 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford(222) as a colourless oil (3.20 g, 94%) which solidified on standing and had m.p. 52 - 53°C after recrystallisation from petroleum ether (b.p. 40 - 60°C).
ÔH (300 MHz, CDCI3): 1.59 - 2.10 (series of m, lOH), 4.40 (m, IH, a-N), 4.51 (m, IH, a-N), 5.19 (s, 2H), 7.30 (m, 5H).
Sc (75 MHz, CDCI3), Signals shown in italics were broadened due to slow N-CO rotation: 19.2, 21.2, 21.6, 32.3 & 32.7 (5 x CHg) 5J.3 (CHN), 67.0 (CHzPh), 76.2 (CHO), 127.9 (2 X aryl CH), 128.3 (aryl CH), 136.5 (aryl C), 154.1 (C=0).
Vmax (CH2CI2): 3035w, 2945s, 2875w, 1715s, 1685s, 1495w, 1435br, 1345m, 1325w, 1300m, 1255W, 1210w, 1150w, 1105s, 1085s, 1035 w, 1030w, lOOOw, 925w, 870w cm"\
™/z (%): 261 (M+, 5), 218 (5), 217 (28), 149 (7), 132 (4), 126 (6), 92 (21), 91 (100), 81
(4), 77(5).
C15H19NO3 [M+] requires ™/z 261.1365; observed 261.136.
Found: C, 68.79; H, 7.30; N, 5.33%.C15H19NO3 requires: C, 68.94; H, 7.33; N, 5.36%.
C;f-4-l (BenzyloxvcarbonvDaminolcvcloheptanol (223)A solution of (222) (2.21 g, 8.47 mmol) in dry ethanol (35 ml) was buffered with sodium phosphate (5.24 g, 36.9 mmol) and stirred for 5 min at 0°C. Freshly prepared®^ and powdered 6% sodium amalgam (33 g) was added and stirring was continued for 1 hr. The mixture was then filtered through celite, and the solvent removed under reduced pressure. The residual solution was partitioned between
153
dichloromethane (30 ml) and water (20 ml). The aqueous layer was extracted with further dichloromethane (2 x 25 ml), the organic layers combined and dried over anhydrous magnesium sulphate. After filtration the solvent was removed under vacuum to afford a white solid which was triturated with 1:1 diethyl ether:petroleum ether to afford (223) (2.18 g, 98 %) which had m.p. 70 - 72°C and was used without further purification. This compound was prepared earlier by Kibayashi^^ in 84% yield (m.p. 72 - 74°C) by reaction of cw-4-aminocycloheptanol with benzyl chlorofonnate
and base.
ÔH (90 MHz, CDCI3): 1.05 - 2.45 (series of m, lOH), 2.50 (brs, IH, exch), 3.75 (brm, 2H, a-O & a-N), 5.10 (s including m, 3H, CH2Ph and NH), 7.30 (m, 5H).
C/5'-4-(Methvlamino)cvcloheptanol (224)A 100 ml flame-dried 2-necked flask fitted with a septum cap and reflux condenser was charged with LAH (220 mg, 5.79 mmol). Dry THF (5 ml) was injected and the system was alternately evacuated and purged with nitrogen gas. The slurry was cooled with stirring to 0°C and a solution of (223) (970 mg, 3.69 mmol) in dry THF (19 ml) was introduced. The mixture was heated at reflux for 2 hr, cooled to 0°C and the minimum of water-saturated diethyl ether was added carefully to destroy excess hydride. The suspension was dried with anhydrous sodium sulphate, filtered through celite and the inorganic residues washed with ethyl acetate ( 2 x 7 ml). The solvent was removed under reduced pressure from the combined organic extracts to give an oil. This was flash chromatographed (to remove benzyl alcohol) using 1:4 methanol:dichloromethane saturated with ammonia, affording (224) (495 mg, 94%) as
a colourless oil which was identical to a sample prepared^^ by hydrogenation of (226). Partition between organic solvents and aqueous solution was avoided since it led to loss of material into the aqueous layer.
5h (90 MHz, CDCI3): 1.80 (m, lOH), 2.45 (s, 3H), 2.70 (m, IH, a-N), 3.25 (s, 2H,
exch), 3.90 (m, IH, a-O).
N-Methvl-8-azabicyclol3.2.1 loctan-l-ol (physoperuvine) (170 ^ 171)A solution of (224) (97 mg, 0.68 mmol) was oxidised using an identical procedure to that described for the preparation of (189 ^ 190) from (192). After removal of acetone under vacuum the residual green solution was dissolved in water (6 ml) and basified to pH 12 using IM sodium hydroxide solution. The water was evaporated
154
under reduced pressure and the residue was extracted continuously with chloroform (25 ml) for 18 hr using a Soxhlet apparatus. After removal of the chloroform under vacuum, (170 ^ 171) (90 mg, 94%) was obtained as a white solid which had m.p. 73 - 74°C after recrystallisation from petrol (b.p. 60 - 80°C) (lit.® m.p. 75°C).
ÔH (300 MHz, 289 K, CDCI3): 1.09 (m, IH), 1.47 (m, 2H), 1.68 (m, 2H), 1.85 (m, 3H), 2.01 (m, 2H), 2.36 (s, 3H), 3.20 (m, IH, a-N), 5.28 (brs, IH, exch).
5c (75 MHz, 298 K, CD2CI2), Signals shown in italics were broadened: 18.3, 23.0 &25.4 (3 X CH2), 29.9 (CH3), 30.3 (CH2), 35.9 (CH2), 58.8 (CHN); the COH signal was not visible at this temperature due to coalescence.
ÔH (300 MHz, 223 K, CD2CI2): 0.86 (brd, J = 12.5 Hz, IH), 1.12 (brd, J = 12 Hz, IH), 1.37 (brm, IH), 1.47 - 2.03 (series of m, 7H), 2.24 (s, 3H), 3.14 (brd, J « 6 Hz, a-N), 6.65 (brs, IH, exch).
Sc (75 MHz, 223 K, CD2CI2), Bicyclic tautomer: 18.3, 21.1, 25.4 & 27.8 (4 x CH2),29.2 (CH3), 36.1 (CH2), 58.1 (CHN), 88.8 (COH).
Monocyclic tautomer (low intensity signals; some were broadened and not all signals were visible): 19.6, 30.9, 35.2, 39.0, 63.3. The ratio of major:minor tautomers was estimated to be approximately 98:2 from integration of the ^^C NMR signals.
Vmax (CH2CI2): 3570W, 3100br, 2940s, 2850m, 2790w, 1695w, 1475m, 1445w, 1325m, 1195m, 1180w, 1145w, 1115m, llOOw, 1040w, 1010m, 975w, 955w, 925w, 910w, 870m, 805w cm"\
*"/z (%): 141 (M+, 65), 140 (7), 113 (60), 112 (70), 99 (57), 98 (85), 84 (36), 71 (21), 70 (100).
CgHjgNO [M" ] requires: *”/z 141.1154; observed 141.1153.
4-(rBenzvloxvcarbonvl1amino)cycloheptanone (225 ^ 226)A solution of (223) (1.02 g, 3.88 mmol) in acetone (18 ml) was oxidised with Jones reagent® using the procedure described for the oxidation of (132) into (133). After work-up, the oil was purified by flash chromatography using 4:1 diethyl
155
ether:petroleum ether (b.p. 40 - 60°C) to give (225 ^ 226) (975 mg, 96%) as a waxy solid.
ÔH (300 MHz, CDCI3): 1.41 (m, IH), 1.60 (m, 2H), 1.81 (m, IH), 2.03 (m, 2H), 2.46 (m, 4H), 3.70 (m, IH, a-N), 5.06 (s, 2H, CHgPh), 5.35 (brd J = 7.1 Hz, IH, NH), 7.31 (s, 5H).
5c (75 MHz, CDCI3): 20.5, 30.4, 36.2, 39.4 & 43.4 (5 x CHg), 52.7 (CHN), 66.5 (CHgPh), 128.0 (2 X aryl CH), 128.4 (aryl CH), 136.5 (aryl C), 155.4 (NC=0), 213.7 (C=0).
Vmax (CH2CI2): 3440m, 3340br, 3030w, 2930m, 2860w, 1705s, 1500s, 1450m, 1405w, 1365w, 1340w, 1310m, 1215s, 1115m, 1065w, 1035w, 1005m, 910w, 875w cm \
% (Cl, %): 262 (MH+, 46), 218 (34), 200 (21), 171 (100), 170 (23), 154 (82), 127 (24), 108 (40), 98 (21), 91 (22), 84 (25).
C15H20NO3 [MH^] requires "'/z 262.1443; observed 262.1443.
1 -Hvdroxv-8-azabicyclo[3.2. Hoctan-1 -ol (norphvsoperuvine) (209 ^ 210)A solution of (225 ^ 226) (593 mg, 2.27 mmol) in absolute ethanol (24 ml) was hydrogenolysed using the procedure described for the reduction of (133) into (176 ^ 177). The solvent was removed under reduced pressure to give a sample of good purity as shown by ^H NMR (265 mg, 92%). Recrystallisation from toluene and
petroleum ether (40 - 60 °C) afforded (209 ^ 210) (225 mg, 78%) as a pale yellow solid, m.p. 97 - 101°C.
5h (300 MHz, 298 K, CDCI3): 1.41 (m, IH), 1.56 (m, 2H), 1.75 (m, 2H), 1.97 (m, 5H), 3.39 (m, IH, a-N), 3.86 (brm, 2H, exch, NH & OH).
5c (75 MHz, 298 K, CDCI3), Signals shown in italics were broadened): 19.3, 28.6, 33.7, 36.0 & 39.8 (5 x CH2), 53.8 (CHN), the COH signal was not resolved at this temperature.
5h (300 MHz, 223 K, CDCI3): 1.35 - 1.97 (series of m, 9H), 2.12 (m, IH), 3.52 (brd, J = 5.4 Hz, IH, a-N).
156
5c (75 MHz, 223 K, CDCI3): 18.7, 26.9, 31.5, 35.0 & 38.7 (5 x CHj), 53.3 (CRN),89.5 (COH).
Vmax (CH2CI2): 36 6 0 W , 35 7 0 W , 3 0 8 0 b r , 2 9 3 0 s , 2 8 6 0 m , 1 6 9 5 m , 1590brw, 1 4 4 5 w ,
1 3 7 5 w , 1 3 5 0 w , 1 3 2 5 w , 1 3 1 5 w , 1 2 2 5 w , 1 1 8 5 m , 1 1 2 0 m , 1 0 8 5 w , 1 0 2 5 m , 9 8 5 w , 9 5 0 w ,
8 8 0 w cm '^.
% (%): 127 (M+, 81), 99 (80), 98 (76), 85 (30), 84 (67), 82 (7), 71 (16), 70 (100).
C7H13NO [M’*'] requires ™/z 127.0997; observed 127.0996.
C?'.y-4-(rBenzvloxvcarbonyl1amino)cvclohept-2-enol (225)A solution of (221) (3.12 g, 12.0 mmol) in absolute ethanol (75 ml) was stirred at 0°C with sodium phosphate (7.70 g, 54.2 mmol) for 5 min. Freshly prepared® and
powdered 6% sodium amalgam (44 g) was added and stining was continued at 0°C for 30 min. The mixture was filtered through celite and the bulk of the solvent distilled under reduced pressure. The residual solution was partitioned between water (40 ml) and dichloromethane (70 ml). The aqueous layer was extracted with further CH2CI2 (2 X 40 ml) and the organic layers combined and dried over anhydrous magnesium sulphate. Filtration and evaporation of solvent left a solid which was recrystallised from ethanol to afford (225) (2.82 g, 90%) as a white solid, m.p. 143 - 144°C. The spectroscopic properties were identical to those of a sample prepared earlier by a different route^^ where it was obtained in 95% yield from cw-4-amino- cyclohept-2-enol by reaction with base followed by benzyl chloroformate, and isolated as an oil.
5h (90 MHz, CDCI3): 1.10 - 2.20 (brm, 6H), 4.30 (brm, 2H, a-N & a-0), 5.10 (s including m, 3H, CH2Ph & NH), 5.50 (brd, J = 12 Hz, IH), 5.80 (brd, J = 12 Hz, IH), 7.30 (m, 5H).
Found C, 69.11; H, 7.14; N, 5.28%.C15H19NO3 requires: C, 68.94; H, 7.33; N, 5.36%.
Ci.y-4-(Methvlamino)cvclohept-2-enol (228)
This compound^^ was prepared by reduction of (227) with LAH using the method
157
described above for (224). The yield was improved to 95% by working up the reaction mixture using water-saturated diethyl ether to destroy the excess hydride. The ethereal solution was then dried with magnesium sulphate, filtered, evaporated,
and the product was flash chromatographed using methanohdichloromethane (1:19) saturated with ammonia gas. The oily product solidified on standing. As in the
prepaiation of (224), partition between organic solvents and aqueous solution was avoided since it led to losses.
ÔH (90 MHz, CDCI3): 2.60 (m, 6H), 2.30 (s, 3H), 3.05 (m, IH, a-N), 3.80 (brs, 2H, exch), 4.15 (m, IH, a-O), 5.60 (dd, J = 11, 6 Hz, IH), 5.90 (dd, J = 11, 6 Hz, IH).
4-r(Benzvloxvcarbonvl)aminolcyclohept-2-enone (229 ^ 230)A solution of (225) (792 mg, 3.03 mmol) in dichloromethane (55 ml) was stirred at room temperature. Barium manganate^^ (8.50 g, 33.2 mmol) was added and stirring was continued for a further 36 hr. The slurry was filtered through celite and the inorganic residues were washed firstly with dichloromethane (2 x 15 ml) and then with warm ethyl acetate (2 x 15 ml). The solutions were combined and the solvent was removed under reduced pressure. The residual oil was purified by flash chromatography using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford
(229 230) as a pale yellow solid (657 mg, 84%), which had m.p. 56 - 58°C afterrecrystallisation from toluene and petroleum ether (b.p. 40 - 60°C).
5h (300 MHz, CDCI3): 1.77 (m, 3H), 2.11 (m, IH), 2.54 (m, 2H), 4.53 (m, IH), 5.08 (s, 2H, CHgPh), 5.69 (brd, J « 8.0 Hz, IH, NH), 5.92 (dd, J = 12.4, 2.2 Hz, IH), 6.34 (dd, J = 12.4, 2.9 Hz, IH), 7.31 (s, 5 H).
ÔC (75 MHz, CDCI3): 18.9, 32.9 & 42.9 (3 x CHg), 51.6 (CHN), 66.8 (CH^Ph), 128.0,128.1 & 128.5 (3 X aryl CH), 130.9 (=CH), 136.2 (aryl C), 146.7 (=CH), 155.6 (NC=0), 202.9 (C=0).
Vmax (CH2CI2): 3440m, 3325br, 3030w, 2960m, 2870w, 1715s, 1665s, 1585w, 1500s, 1450m, 1215s, 1155w, 1125w, 1050w, 1020m, 980w, 900w, 795w cm'^
™/z (%): 259 (M+, 4), 241 (3), 215 (18), 198 (6), 172 (12), 168 (16), 159 (21), 140 (6), 124(17), 110 (66), 91 (100).
158
Found: C, 69.49; H, 6.52; N, 5.19%;C15H17O3N requires: C, 69.48; H, 6.61; N, 5.40%.
4-(Methylamino')evclohept-2-enone (231 ^ 232)A solution of the amino-alcohol (226) (48 mg, 0.34 mmol) in acetone (7 ml) was
acidified with trifluoroacetic acid (30 pi, 0.39 mmol) at ambient temperature. The solution was titrated with Jones reagent^^ and stirred for a further 5 min before the addition of excess isopropanol. The solution was filtered and the solid washed with methanol ( 2x5 ml). The organic solyent was distilled off under reduced pressure and the residual oil was dissoWed in water (5 ml), and the solution was basified to pH 12 using aqueous sodium hydroxide solution (2M). The solution was extracted with ethyl acetate ( 3 x 7 ml), the organic layers were combined, dried oyer anhydrous magnesium sulphate, filtered, and the solyent distilled under yacuum to afford the amino-ketone (231 ^ 232) as an oil (19 mg, 40%). The compound decomposed during chromatography. Salts of the amine were not easy to handle, for example the HBF4 salt sepaiated from dry diethyl solution as an oil and a pure sample was not obtained. The free amine decomposed oyer a period of hours.
Sh (300 MHz, CDCI3): 1.75 - 2.76 (series of m, 6H), 2.59 (s, 3H), 3.74 (s, IH, a-N), 6.01 (dd, J = 10.2,0.9 Hz, IH), 6.35 (dd, J = 10.2, 3.2 Hz, IH).
Vmax (CH2CI2): 3040w, 2920m, 2840w, 2790w, 1665m) with shoulders at 1755w and 1700m, 1425w, 1200s, 1180m, 1140m, 905s cm'^
% (%) [measured using (3:HC1)]: 140 (MH+, 32), 139 (M+, 100), 124 (22), 113 (43), 112 (43), 111 (54) 110 (40), 98 (43), 96 (28), 83 (98), 82 (58), 70 (62), 69 (27), 68 (97), 57 (63), 55 (91).
4-(Benzylamino)cvclohept-2-enone (233 234)A solution of (56) (84 mg, 0.39 mmol) in acetone (12 ml) was cooled to 0“C and acidified with trifluoroethanoic acid (60 pi, 0.62 mmol). The solution was titrated with Jones reagent® until an orange/brown colour persisted and excess oxidant destroyed immediately by addition of excess isopropanol. The solution was filtered through celite and the remaining solid washed with acetonee ( 3 x 4 ml). The bulk of the solvent was distilled under reduced pressure and the resulting green oil was dissolved in water (3 ml) and basified to pH 12 with sodium hydroxide solution (2M).
159
The solution was extracted with ethyl acetate (3 x 10 ml), the organic layers were combined, washed with brine (5 ml) and dried over anhydrous magnesium sulphate. Separation, filtration and distillation of solvent under reduced pressure afforded (233
234) as a pale yellow oil (71 mg, 85%) which was approximately 90% pure (from NMR integration) but decomposed on standing in solution.
ÔH (300 MHz, CDCI3): 1.75 (m, 3H), 2.11 (m, IH), 2.52 (m, 2H), 3.55 (m, a-N, IH),
3.84 (ABq, J= 13.1 Hz, 2H, CH^Ph), 5.97 (ddd, J = 12.3, 2.3, 1.0 Hz, IH), 6.56 (ddd, J = 12.3, 3.7, 1.0 Hz, IH), 7.32 (m, 5 H).Double irradiation at Ô 2.52 caused the ddd at Ô 5.97 to simplify to a dd (J = 12.3, 2.3 Hz). Double irradiation at 5 3.55 caused the ddd at S 5.97 to simplify to a doublet (J =12.3 Hz) and the ddd at 5 6.56 to simplify to a dd (J = 12.3, 1.0 Hz).
ÔC (75 MHz) 19.1, 32.2 & 42.7 (3 x CHg), 51.5 (CH^Ph), 56.7 (CHN), 127.1, 128.1 &128.5 (3 X aryl CH), 130.7 (=CH), 139.5 (aryl C), 149.5 (=CH); the cycloheptenone carbonyl signal was observed only at 233 K (0^ 204.8). The ^^C NMR spectrum at 233K showed additional signals but the peaks could not be fully assigned; pronounced broadening of the minor signals was probably associated with slow inversion at nitrogen in the bicyclic tautomer.
Vmax (CH2CI2): 3380brw, 3030w, 2940s, 2870m, 1665s (with shoulders at 1755m and 1705m), 1610W, 1495 w, 1455m, 1395w, 1340w, 1200m, 1140w, 1105w, 1070w, 1030w, 980w, 910s cm"\
™/z (%) [measured using (233 ^ 234.HC1)]: 216 (MH+, 14), 215 (M+, 16), 133 (11), 126 (19), (121 (14), 106 (16), 105 (45), 104 (11), 98 (45), 92 (16), 91 (100), 77 (25).
C14H17NO requires 215.1310; observed 215.1309.
Attempts to obtain a salt with anhydrous HBF4 in diethyl ether gave only a gum. Passage of anhydrous HCl through a diethyl ether solution of the amino-ketone gave a
salt which could not be recrystallised but was partially purified by trituration with anhydrous diethyl ether; it was hygroscopic and deteriorated on standing in air.
C/.y-4-[(Benzvloxvcaibonvl)aminol-l-[(p-toluenesulphonvl)oxvcvclooct-2-ene (236)A solution of n-butyllithium (2.5M in hexane, 2.69 ml, 6.72 mmol) was injected into a
160
stilled solution of (150) (1.683 g, 6.23 mmol) in THF (28 ml) at -78°C under a nitrogen atmosphere. After 10 min a solution of p-toluenesulphonyl chloride (1.510 g, 7.91 mmol) in THF (8 ml) was added and warmed to room temperature. After stirring for 1 hr the reaction was quenched with water (1 ml) and the bulk of the solvent was distilled under reduced pressure. The residual oil was dissolved in diethyl ether (45 ml), dried over anhydrous magnesium sulphate, filtered and the solvent evaporated under reduced pressure. The crude oil was purified by flash chromatography using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) to yield (236) (1.740 g, 66%) as a yellow oil. Some decomposition occured during chromatography and also on standing. Full characterisation was not attempted and the tosylate was used immediately in the next reaction.
5h (300 MHz, CDCI3): 1.27 - 2.08 (series of m, 8H), 2.39 (s, 3H), 3.95 (m, IH), 4.29 (m, IH), 4.86 (brd, J = 5.4 Hz, NH), 5.11 (s, 2H, CHjPh), 5.28 (m, IH), 5.40 (dd, J =10.4, 6.9 Hz, IH), 7.27 (part of AA’BB’, 2H), 7.36 (m, 5H), 7.81 (part of AA’BB’, 2H).
7>an5'/c/i'-4-r(Benzvloxvcaibonvl)aminol-l-chlorocvclooct-2-ene (237)Lithium chloride (1.102 g, 26 mmol) was dissolved in DMSO (16 ml) and warmed with stirring to 60°C. A solution of (236) (1.610 g, 3.75 mmol) was added and stirred for 45 min. The solution was poured into water (15 ml) and extracted with diethyl ether (4 x 20 ml). The combined organic layers were washed with water (7 ml) and brine (5 ml). After drying with anhydrous magnesium sulphate and filtration, the
solvent was distilled under reduced pressure. The crude oil was purified by flash chromatography eluting with 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C). The first fraction to be eluted (87 mg) was unidentifiable, but further elution afforded (237) (489 mg, 45%) as a white solid which had m.p. 87 - 88°C after recrystallisation from ethanol. The ^H NMR indicated a 7:3 trans'.cis mixture of isomers from signal integration; figures quoted in italics are common to both isomers.
Sh (300 MHz, CDCI3): 1.34 (m, IH), 1.45 -1.74 (series of m, 3H), 1.86 (m, 2H), 2.13 (m, 2H), 4.44 (brm, IH), 4.78 (brm, IH), 4.91 (brm, IH), 5.08 (s, 2H, CHgPh), 5.31 (ddd, J = 10.7, 8.1, 1.3 Hz, IH, trans-isomer), 5.38 (dd, J = 12.4, 6.1, IH, cw-isomer), 5.67 (ddd, J = 10.7, 8.0, 1.4 Hz, IH, trans-isomer), 5.76 (dd, J = 12.1, 6.1 Hz, IH, cw-isomer), 7.33 (m, 5H).
161
5c (75 MHz, CDCI3), Tran^-isomer: 23.5, 24.9, 36.7 & 40.1 (4 x CHg), 49.3 (CHN),56.7 (CHCl), 6 6 . 6 (CHzPh), 128.0 (aryl CH), 128.4 (2 x CH), 130.9 (czCHg), 132.1 (=CHg), 136.2 (aryl C), 155.4 (C=0).
Cw-isomer: 22.7, 23.9, 33.6 & 36.5 (4 x CH^), 49.2 (CHN), 57.9 (CHCl), 6 6 . 6
(CH2PI1), 128.0 (aryl CH), 128.4 (2 x CH), 130.9 (=€«2), 132.1 (=CH2), 136.2 (aryl C), 155.4 (C=0).
^max (CH2CI2): 3440m, 3340w, 3030w, 2940m, 2860w, 1715s, 1505s, 1465w, 1450m, 1395W, 1370W, 1325m, 1215s, 1130w, 1085w, 1025m, 980w, 910w, 800w, 790w cm'^
% (%): 295 (M+, 0.5), 293 (M+, 2), 259 (18), 258 (100), 249 (8), 214 (37), 204 (4),202 (11), 197 (9), 172 (41), 166 (9), 108 (10).
Found: C, 65.12; H, 6.67; N, 4.76%.C16H20NO2CI requires: C, 65.40; H, 6.86; N, 4.77%.
N-(Benzvloxvcarbonvl)-9-azabicyclor4.2. llnon-7-ene (238)Sodium hydride (60% dispersion in mineral oil, 97 mg, 2.43 mmol) was slurried with THF (2 ml) under a nitrogen atmosphere. A solution of (237) (341 mg, 1.16 mmol) in THF (8 ml) was injected via syringe and stirred at ambient temperature for 1.5 hr and then refluxed for a further 2.5 hr. Excess hydride was destroyed by quenching with the minimum quantity of water at -78°C. The bulk of the solvent was distilled under reduced pressure and the oily residue was paititioned between diethyl ether (18 ml) and water (7 ml). After washing with further water (5 ml), the organic layer was sepaiated, dried with anhydrous magnesium sulphate, filtered and the solvent distilled under vacuum. Purification by flash chromatography using 1:4 diethyl ether: petroleum ether (b.p. 40 - 60°C) afforded firstly the aziridine (239) (31 mg, 10%) as a pale yellow oil.
Sjj (300 MHz, CDCI3), Double irradiation of the two aziridine protons allowed for full measurement of the coupling constants: 1.22 - 1.44 (m, 2H), 1.51 - 1.84 (series of m, 3H), 1.99 (m, IH), 2.15 (dddd, J = 13.8, 5.8, 3.4, J « 3 Hz, IH), 2.28 (m, IH), 2.63 (ddd, J = 10.1, 6.3, 3.4 Hz, IH, a-N), 3.07 (ddd, J = 6.3, 1.7, 1.1 Hz, IH, a-N), 5.13 (s, 2H, CHzPh), 5.57 (dd, J = 11.1, 1.1 Hz, IH, HC=), 5.75 (dddd, J = 11.1, 7.6, 5.8,
162
1.7 Hz, IH, HC=), 7.35 (m, 5H).
ÔC (75 MHz, CDCI3): 25.7, 26.7, 27.1 & 28.8 (4 x CHj), 40.0 & 44.4 (2 x CHN), 68.0 (CH2PI1), 122.3 (=CH2), 128.1, 128.2 & 128.5 (3 x aiyl CH), 134.9 (=CH), 136.0 (aryl C), 163.8 (C=0).
Pmax (CH2CI2): 3030W, 2940m, 2860w, 1715s, 1495w, 1450w, 1425w, 1380w, 1270brs, 1215s, 1170w, lllOw, 1090w, 1045w, 1025w, 910wcm'^
% (%): 257 (M+, 1), 171 (10), 170 (16), 122 (14), 92 (9), 91 (100).
C16H19NO2 [M+] requires '"/g 257.1416; observed 257.1416.
Further elution with 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C) yielded (238) (86 mg, 28%) as a colourless oil; the signals quoted below in italics are common to both rotamers (in a 1:1 ratio).
ÔH (300 MHz, CDCI3): L27 -1.66 (series of m, 6H), 1.94 (m, IH), 2.07 (m, IH), 4.72 (brd, J = 6.1 Hz, IH, a-N), 4.78 (brd, J = 6.0 Hz, IH, a-N), 5.13 & 5.18 (ABq, J =12.5 Hz, 2H, CH2Ph), 5.72 (dd, J = 6.8, 2.3 Hz, IH), 5.75 (dd, J = 6.8, 2.3 Hz, IH),7.35 (m, 5H).
ÔC (75 MHz, CDCI3): 24.0 & 24.2 (2 x CH2), 30.8 &31.7 (2x CH2), 60.9 &61.3 (2 x CHN), 127.7, 127.8 & 128.4 (3 x aryl CH), 130.9 & 131.0 (2 x =CH), 137.1 (aryl C),153.2 (C=0).
Pniax (CH2CI2): 3030w, 2930s, 2860m, 1695s, 1615w, 1585w, 1495m, 1430s, 1365m, 1350m, 1315s, 1265brw, 1230w, 1210w, 1195w, 1180w, 1140m, 1115s, 1100s, 1095s, 1070s, 1025w, 975m, 960w cm'^.
"^4 (%): 257 (M+, 11), 171 (12), 170 (21), 92 (9), 91 (100).
C16H19NO2 [M'*'] requires "'/g 257.1416; observed 257.1418.
In a repeat experiment, the yield of (238) was increased to 40% by using a 6:1 mixture of THF:DME as solvent and stirring at 60°C for 2.5 hr, rather than refiuxing.
163
The yield of aziridine (239) was suppressed to 3% using these conditions.
N-(Benzvloxvcai‘bonvl)-73,83-epoxv-9-azabicycIor4.2.11nonane (235)A solution of (238) (39 mg, 0.15 mmol) in dichloromethane (4 ml) was epoxidised with MCPBA (50 - 60% purity, 57 mg, 0.18 mmol) using an identical procedure to
that described for the conversion of (151 ^ 152) into (194). Purification of the crude product by flash chromatography using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) afforded (235) (34 mg, 84%) as a colourless oil. The chemical shifts quoted in italics refer to signals common to both rotamers (in a 1:1 ratio).
ÔH (300 MHz, CDCI3): 1.48 (m, 6H), 1.90 (m, IH), 2.03 (m, IH), 3.36 (d, J = 3.0 Hz,
IH, HCO), 3.37 (d, J = 3.0 Hz, IH, HCO), 4.35 (d, J = 6.2 Hz, IH, a-N), 4.41 (d, J =6.2 Hz, IH, a-N), 7.33 (m, 5H).
5c (75 MHz, CDCI3): 24.3 (2 x CH2), 28.8 & 29.6 (2 x CH2), 55.9 & 56.0 (2 x CHO),56.8 & 57.3 (2 x CHN), 66.7 (CHzPh), 127.7, 127.9 & 128.4 (3 x aryl CH), 136.8 (aryl C), 155.1 (C=0).
Pjnax (CH2CI2): 3040w, 2930m, 2860w, 1695s, 1495w, 1425s, 1350w, 1320m, 1295w, 1195w, 1155w, 1115m, 1095m, 1035w, 1025w, 980w, 900w, 850m cm'^
% (%): 273 (M+, 22), 166 (13), 138 (8), 110 (5), 92 (8), 91 (100).
C16H19NO3 [M+] requires 273.1365; observed 273.1363.
N-(Benzvloxvcaibonvl)-73-hvdroxv-9-azabicvclor4.2.Hnonane (240)A solution of (238) (78 mg, 0.30 mmol) in THF (4 ml) was cooled with stirring to
-78°C under a nitrogen atmosphere. Borane:THF complex (IM in THF, 0.18 ml, 0.18 mmol) was injected and the solution was slowly warmed to room temperature. After2.5 hr the reaction was quenched by the sequential addition of water (200 pi), sodium hydroxide solution (6M, 200 pi) and hydrogen peroxide solution (30 weight %, 200 pi). The reaction mixture was stirred for a further 10 min, the bulk of the solvent distilled under reduced pressure and the residue partitioned between diethyl ether (20 ml) and water (5 ml) and then washed with further water (5 ml) and brine (5 ml). The diethyl ether layer was dried over anhydrous magnesium sulphate, filtered and the solvent distilled in vacuum. The resultant crude oil was purified by flash
164
chromatography using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford(240) (59 mg, 71%) as a colourless oil. The chemical shifts quoted in italics refer to signals common to both rotamers.
ÔH (300 MHz, CDCI3): 1.24 - 1.64 (m, 4H), 1.81 - 2.16 (series of m, 6H), 2.89 (brs, IH, exch), 4.17 (m, IH), 4.49 (m, IH), 5.05 & 5.18 (ABq, J = 12.4 Hz, 2H, CHjPh), 7.33 (s, 5H).
Sc (75 MHz, CDCI3): 23.8, 23.9, 24.0 & 24.1 (4 x CH2), 30.5, 31.5, 32.0 & 33.1 (4 x
CH2), 40.1 & 40.7 (2 X CHg), 55.2 & 55.8 (2 x CHN), 65.2 & 65.4 (2 x CHN), 66.5 &66.6 (2 X CHzPh), 77.4 & 78.4 (2 x CHOH), 127.4 & 127.5 (2 x aryl CH), 127.8 (aryl CH), 128.3 (aryl CH), 136.7 & 136.8 (2 x aryl C), 154.3 & 154.5 (2 x C=0).
Umax (CH2CI2): 3600w, 3460brw, 3040w, 2940w, 2860w, 1690s, 1500w, 1445w, 1420m, 1360w, 1335m, 1215w, 1190w, 1115m, 1095m, 1020w, lOOOw, 965w, 940w, 910 cm'K
«14 (%): 275 (M+, 17), 184 (11), 168 (5), 141 (4), 140 (40), 96 (20), 91 (100).
C16H21NO3 [M+] requires *"4 275.1521; observed 275.1524.
Norhomotropan-7 3-ol (241)The carbamate (240) (49 mg, 0.18 mmol) was hydrogenolysed in absolute ethanol (5 ml) using the procedure described for the conversion of (194) into (197). The amine(241) (24 mg, 96%) was isolated as a thick oil after distillation of solvent under vacuum.
Sjj (300 MHz, CD3OD): 1.38 - 1.81 (series of m, 8H), 1.88 (ddd, J = 14.5, 8.7, 2.8 Hz), 1.97 (ddd, J = 14.5, 6.2, 3.3 Hz), 3.30 (brd overlapping with solvent signal, IH, a-N), 3.77 (m, IH, a-N), 4.12 (dd, J = 6.2, 2.8 Hz, IH, a-OH).
Sc (75 MHz, CDCI3): 24.3, 24.7, 33.4, 35.4 & 42.2 (5 x CHj), 57.3 & 66.5 (2 x CHN),79.8 (CHOH).
Umax (CH2CI2): 3600w, 3300brw, 2930s, 2860w, 1460w, 1440w, 1255w, llSOw, 1120w, 1050w, 995w, 950w, 905m cm’^
165
""/g (%): 141 (M+, 31), 124 (31), 98 (33), 97 (100), 96 (17), 84 (17), 82 (51), 70 (11), 69 (30), 68 (36).
CgH^gNO [M+] requires ™/g 141.1154; observed 141.1153.
l3-Hvdroxv-2B,3B-epoxv-4B-r(benzvloxYcarbonvl)aiiiino1cvclooctane (244)Using the procedure of Soai;^^ methanol (300 pi) was added over 45 min via a syringe pump to a refiuxing solution of (194) (126 mg, 0.44 mmol) and sodium borohydride (42 mg, 1.11 mmol) and the solution was then refluxed for 2 hr. After cooling to room temperature, saturated ammonium chloride solution (1 ml) was added
to destroy excess hydride. Diethyl ether (20 ml) was added and washed with saturated ammonium chloride solution (12 ml). The ethereal layer was separated, dried with anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure. The residual oil was purified by flash chromatography using 3:2 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford (244) (92 mg, 73%) as a colourless oil. In subsequent preparations the yield of (244) could be improved to 98% by refiuxing in THF for 3 hr with 2 M equivalents of sodium borohydride (without the addition of methanol). The product isolated using this procedure was pure enough for further reactions without chromatography.
ÔH (300 MHz, CDCI3): 1.38 (m, 2H), 1.66 (m, 4H),1.86 (m, 2H), 2.91 (dd, J = 4.5, 3.6 Hz, IH, HCO), 3.11 (t, J = 4.5, J « 4.8 Hz, IH, HCO), 3.62 (brs, IH, exch), 4.42 (brd, J « 3.4 Hz, IH), 4.47 (m, IH), 5.06 (s, 2H, CH^Ph), 7.15 (d, J = 9.3 Hz, IH, NH), 7.31 (m, 5H).
Sc (75 MHz, CDCI3): 22.0, 22.9, 31.3 & 32.0 (4 x CHg), 46.5 (CHN), 57.1 & 58.6 (2 X CHO), 65.5 (CHOH), 66.4 (CHjPh), 127.8, 127.9 & 128.3 (3 x aryl CH), 136.8 (aryl C), 156.2 (C=0).
Umax (CH2CI2): 3570brw, 3330brw, 3030w, 2940m, 2870w, 1710s, 1525m, 1450w, 1380w, 1340w, 1305w, 1235m, 1165w, 1130w, 1085w, 1070w, 1040m, 1025w, 1005w, 925w, 905m cm"\
™/g (%): 291 (M+, 3), 156 (6), 146 (5), 108 (13), 91 (100).
166
C16H21NO4 [M+] requires '"/g 291.1471; observed 291.1474.
The above reaction was reversible using the previously described Jones procedure used to oxidise (132); a solution of (244) (34 mg, 0.12 mmol) was oxidised to (243) (32 mg, 95%) which appeared pure from the 90 MHz ^H NMR spectrum.
lB-r(p-Toluenesulphonvl)oxv1-2B.3B-epoxv-4B-r(benzvloxvcarbonvl)aminol- cyclooctane (245)
A solution of (244) (520 mg, 1.78 mmol) in dry THF (11 ml) was stirred at 0°C under a nitrogen atmosphere. A solution of n-butyllithium (2.5M in hexane, 0.86 ml, 2.15 mmol) was injected and stirred for 5 min, before the addition of p-toluenesulphonyl chloride (443 mg, 2.32 mmol) in THF (4 ml). The solution was warmed to room temperature and stirred a further 1.5 hr and then quenched with water (0.5 ml). Diethyl ether (30 ml) was added, the solution transferred to a separating funnel, and washed with water ( 2 x 7 ml) and brine (7 ml). After separation the ethereal layer was dried over anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure. The tosylate (245) (728 mg, 92%) was isolated as a foam by purification of the crude oil using flash chromatography, eluting with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C).
Sh (300 MHz, CDCI3): 1.08 - 1.36 (brm, 2H), 1.44 - 1.73 (series of m, 5H), 2.02 (m, IH), 2.42 (s, 3H, Me), 3.11 (m, 2H, CHO), 4.11 (m, IH, a-N), 4.91 (brd, J = 9.1 Hz, IH, a-0 S02Ar), 5.09 (s, 2H, CH2Ph), 5.41 (brd, J = 8.8 Hz, IH, NH), 7.29 - 7.33 (s, 5H plus, part of AA’BB’, 2H), 7.80 (part of AA’BB’, 2H).
Sc (75 MHz, CDCI3): 21.7 (CH3), 22.3, 23.3, 27.3 & 27.4 (4 x CHg), 49.3 (CHN),59.1 & 59.5 (2 X CHO), 66.8 (CH^Ph), 79.4 (CHOSOg), 127.8, 128.0, 128.1, 128.5 &130.0 (5 X aryl CH), 133.7 (aryl C), 136.4 (aryl CCHj), 145.0 (aryl CSO3), 155.6 (C=0).
P m a x ( C H 2 C I 2 ) : 3440W, 3030w, 2940W, 2870w, 1720s, 1600w, 1505m, 1455w, 1365m, 1220w, 1190m, 1185s, 1095w, 1045w, 1025w, 900m, 855m, 810w cm '\
™/g (%): 445 (M+,9), 395 (23), 377 (21), 363 (74), 352 (21), 345 (58), 335 (61), 329 (59), 320 (100).
167
C23H27NO6 [M+] requires ""/g 445.1559; observed 445.1558.
1 q-Chloro-2B,3 3-epoxv-4B-r(benzvloxvcarbonvl')aminolcvclooctane (246)Lithium chloride (870 mg, 21 mmol) and (245) (1.538 g, 3.46 mmol) were added to
DMSO (17 ml) and heated to 60°C with stirring for 1.5 hr. The solution was poured into an equal volume of water and repeatedly extracted with diethyl ether (3 x 25 ml). The organic layers were combined and washed with water ( 2 x 7 ml) and brine (5 ml), before drying over anhydrous magnesium sulphate. Filtration and distillation of solvent under reduced pressure gave a crude oil which was purified by flash chromatography using 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford (246) (928 mg, 87%) as a white solid. An analytical sample was prepared by recrystallisation from ethanol (m.p. 94 - 95°C).
5h (300 MHz, CDCI3): 1.43 (m, 2H), 1.58 (m, IH), 1.66 - 1.92 (series of m, 3H), 2.04 (m, 2H), 3.14 (dd, J = 8.9, 4.3 Hz, HCO, B-Cl), 3.37 (t, J » 4.3 Hz, HCO, p-N), 4.24 (m, 2H, a-N & a-Cl), 5.08 & 5.12 (ABq, J = 12.2 Hz, 3H, CH2Ph & NH), 7.35 (m, 5H).
5c (75 MHz, CDCI3): 22.6, 24.0, 29.8 & 36.2 (4 x CH2), 49.3 (CHN), 58.6 & 59.1 (2 X CHO), 61.7 (CHCl), 66.9 (CH2Ph), 128.1, 128.2 & 128.5 (3 x aryl CH), 136.2 (aryl C), 155.6 (C=0).
Pmax (CH2CI2): 3680W, 3430w, 2940w, 2860w, 1720s, 1605w, 1500m, 1450w, 1335w, 1310w, 1220m, 1165w, 1130w, 1105w, 1055w, 1035w, 895 cm' .
% (%): 311(M+, 12), 309 (M+, 32), 273 (18), 230 (11), 122(6), 108 (21), 107 (7), 92 (7), 91 (100).
Found: C, 62.17; H, 6.41; N, 4.64%.C16H20NO3CI requires: C, 62.02; H, 6.51; N, 4.52%.
N-(Benzvloxvcarbonvl)-7B,8B-enoxv-9-azabicvclo[4.2.11nonane (235)To a solution of (246) (414 mg, 1.34 mmol) in THF:DME (5:1, 8 ml) was added sodium hydride (60% dispersion in mineral oil, 107 mg, 2.68 mmol) and stirred for 2 hr. The solution was cooled to -78°C and excess hydride was destroyed by the addition of the minimum quantity of water. Diethyl ether (15 ml) was added and
168
washed with water ( 2 x 5 ml) and brine (5 ml). The organic layer was dried with anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure. The crude oil was purified by flash chromatography, eluting with 2:3 diethyl ether:petroleum ether (b.p. 40 - 60°C), to afford (235) (358 mg, 98%) as a colourless oil. The epoxide prepared using this method was identical to a sample prepared by epoxidation of the alkene (238).
lB-Hvdroxv-2a,3a-epoxv-4B-r(benzvloxvcarbonvl)aminolcyclooctanone (255) MCPBA (50 - 60% purity, 4.21 g, 13.4 mmol) was added to stirred solution of (150) (3.07 g, 11.2 mmol) and stirring was continued at room temperature for 3 hr. The solution was transferred to a separating funnel and washed with saturated sodium bicarbonate solution (2 x 20 ml), dried oyer anhydrous magnesium sulphate, filtered and the solyent distilled under vacuum. The residual oil was purified by flash chromatography, eluting with diethyl ether, to afford (255) (3.10 g, 95%) as a white solid which had m.p. 125 - 126°C after recrystallisation from ethanol.
ÔH (300 MHz, CDCI3): 1.39 - 1.70 (series of m, 6H), 1.86 (brm, 2H), 2.95 (dd, J = 4.7,9.3 Hz, IH, HCO), 3.03 (dd, J = 4.7, 8.2 Hz, HCO), 3.56 (m, IH), 3.70 (m, IH), 4.16 (brs, IH, exch), 5.01 (m, IH, NH), 5.29 (s, 2H, CH^Ph), 7.34 (m, 5H).
ÔC (75 MHz, CDCI3): 23.2, 24.4, 33.6 & 34.8 (4 x CHj), 51.3 (CHN), 57.9 & 60.3 (2 X CHO), 66.7 (CHjPh), 71.2 (CHOH), 128.0 (2 x aryl CH), 128.4 (aryl CH), 136.4 (aryl C), 155.7 (C=0).
Pmax (CH2CI2): 3600w, 3430w, 2930m, 2860w, 1720s, 1510m, 1450w, 1315brw, 1225m, 1145w, 1090w, 1045m, 1025w, 965w, 900w cm '\
% (%): 291 (M+, 45), 256 (22), 237 (58), 213 (29), 200 (22), 184 (46), 167 (77), 108(49), 91 (100).
CigH2iN04 [M+] requires "'/g 291.1471; observed 291.1474.
Found: C, 66.28; H, 7.46; N, 5.00%.C16H21NO4 requires: C, 65.96; H, 7.27; N, 4.81%.
169
2a.3a-Epoxy-43-r (benzvloxvcarbonvDaminolcyclooctanone (256)A solution of (255) (551 mg, 1.90 mmol) in acetone (25 ml) was oxidised with Jones reagent using the procedure described preyiously for the conyersion of (132) into (133). Reciystallisation of the crade product with ethanol afforded (256) (523 mg, 96%) as a ciystalline white solid (m.p. 145 - 146°C).
Sh (300 MHz, CDCI3): 1.48 (m, IH), 1.62 - 1.94 (series of m, 5H), 2.35 (ddd, J =13.4, 10.2, 3.7 Hz, IH), 2.63 (m, IH), 3.04 (m, IH, a-N), 3.36 (m, IH, HCO), 3.82 (d, J = 5.2 Hz, IH, HCO), 5.03 (s, 2H, CH^Ph), 5.61 (brd, J = 6.1 Hz, IH, NH), 7.30 (s, 5H).
Sc (75 MHz, CDCI3): 24.2, 24.7, 33.3 & 42.6 (4 x CHj), 52.3 (CHN), 57.5 & 58.5 (2 X CHO), 66.6 (CHzPh), 128.0 (2 x aryl CH), 128.4 (aryl CH), 136.4 (aryl C), 155.5 (NC=0), 206.2 (C=0).
P m a x ( C H 2 C I 2 ) : 34 3 0 W , 3 0 3 0 w , 2 9 4 0 w , 1 7 2 5 s , 1 5 0 5 m , 1 4 5 0 w , 1 3 7 5 w , 1 3 1 5 w ,
1 2 2 0 m , 1 1 8 5 w , 1 1 3 5 w , llOOw, 1 0 7 0 w , 1 0 2 0 m , 9 5 0 w , 9 1 5 w , 8 4 5 w c m l
% (%): 289 (M+,8), 183 (3), 112 (4), 108 (15), 107 (9), 92 (9), 91 (100).
Found; C, 66.46; H, 6.46; N, 5.02%.
C16H19NO4 requires; C, 66.42; H, 6.62; N, 4.84%.
1 a-Hydroxy-2a,3a-epoxy-43-r(benzyloxycarbonyl)aminolcyclooctane (257)A solution of (256) 1.105 g, 3.82 mmol) and sodium borohydride (350 mg, 9.46 mmol) in THF (45 ml) was refluxed for 1.5 hr. Saturated ammonium chloride solution (2 ml) was added to destroy excess hydride and the bulk of the solyent was distilled under reduced pressure. The residual oil was dissolved in diethyl ether (30 ml) and washed with saturated ammonium chloride solution (10 ml) and water (10 ml). The organic layer was dried over anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure. The crude oil was purified by flash chromatography, eluting with 7:3 diethyl ether:petroleum ether (b.p. 40 - 60°C), to afford (257) (982 mg, 88%) as a foam. The foam was triturated to a white solid with petroleum ether (b.p. 40 - 60°C) and an analytical sample was prepared by recrystallisation with ethanol (m.p. 103 - 104 °C). The ^H NMR spectrum was broad when recorded in CDCI3, but signals were better resolved in CD3COCD3.
170
ÔH (300 MHz, CDCI3): 1.20 (m, IH), 1.39 - 1.66 (m, 4H), 1.82 (m, 2H), 2.09 (m, IH), 2.75 (vbrs, IH, exch), 3.04 (brm, 2H, HCO), 4.42 (brm, 2H, a-N & a-OH), 5.08 (m, 3H, CH^Ph & NH), 7.33 (m, 5H).
ÔH (300 MHz, CD3COCD3): 1.34 (m, IH), 1.52 - 1.89 (series of m, 6H), 2.01 (m, IH), 2.93 (dd, J = 9.4, 4.2 Hz, IH, HCO, p-N), 3.06 (t, J = 4.2 Hz, IH, HCO, p-OH), 3.41 (brs, IH, exch), 4.39 (m, IH), 4.52 (brm, IH), 5.05 (s, 2H, CH2Ph), 6.44 (brd, J = 6.8 Hz, NH, IH), 7.35 (m, 5H).
ÔC (75 MHz, CDCI3): 19.3 & 25.0 (2 x CH2), 32.6 (2 x CH2), 48.8 (CHN), 57.2 &59.3 (2 X CHO), 65.4 (CHOH), 66.7 (CH2Ph), 128.1 (2 x aryl CH), 128.5 (aiyl CH),136.5 (aryl C), 156.0 (C=0).
Pmax (CH2CI2): 3560m, 3430m, 3350w, 3040w, 2940s, 2865m, 1720s, 1505s, 1455m, 1405w, 1305w, 1235s, 1220s, 1180w, 1165w, 1140w, 1095s, 1025s, 985w, 930w, 910w, 890m cm"k
% (CI, %): 292 (MH+, 100), 248 (9), 201 (10), 156 (27), 140 (16), 108 (26).
C16H22NO4 [MH+] requires % 292.1549; observed 292.1549.
Found: C, 65.95; H, 7.13; N, 4.72%.C15H19NO4 requires: C, 65.96; H, 7.27; N, 4.81%.
la-r(p-toluenesulphonyl)oxvl-2a,3a-er)oxv-43-r(benzvloxvcaibonvl)aminolcvclo- octane (258)A solution of (257) (307 mg, 1.05 mmol) was tosylated using the procedure described previously for the tosylation of (244). The residual oil was purified by flash chromatography, eluting with 4:1 diethyl ether:petroleum ether (b.p. 40 - 60°C), to afford (258) (451 mg, 96%) as a gum. The ^H NMR spectrum was broad when recorded in CDCI3, but signals were better resolved in CD3COCD3.
ÔH (300 MHz, CD3COCD3): 1.38 - 1.59 (m, 2H), 1.61 - 1.83 (series of m, 5H), 1.98 (m, IH), 2.43 (s, 3H), 2.90 (dd, J = 9.3, 4.2 Hz, IH, HCO, p-N), 3.12 (t, J = 4.2 Hz, IH, HCO, p-OSOjAr), 4.13 (m, IH, a-N), 5.08 (s, 2H, CHgPh), 5.20 (ddd, J = 6.2,
171
3.7, 2.2 Hz, IH, a-OS02Ar), 6.48 (m, IH, NH), 7.38 (m, 5H, plus part of AA’BB’,2H), 7.89 (part of AA’BB’, 2H).
ÔC (75 MHz, CDCI3): 20.7 (CHg), 21.6 (CH3), 23.8, 31.4 & 32.5 (3 x CHg), 49.5 (CHN), 58.4 & 57.7 (2 x CHO), 66.5 (CH^Ph), 75.7 (CHOSOgAr), 127.8 (2 x aryl
CH), 128.1, 128.4 & 129.4 (3 x aryl CH), 133.7 (aryl CMe), 136.5 (aryl CCHg), 144.5 (aryl CSOj), 155.3 (C=0).
Umax (CH2CI2): 3440m, 3040w, 2940m, 2865w, 1725s, 1600w, 1510s, 1455m, 1395w, 1360s, 1325W, 1305w, 1220s, 1190s, 1175s, 1140w, 1120w, 1095m, 1080m, 1040m, 1025m, 1015m, 930s, 880w, 825m, 815m cm \
% (%): 445 (M+, 1), 273 (7), 187 (45), 186 (18), 170 (10), 169 (100), 155 (40), 149(15), 138 (54), 123 (14), 109 (21), 95 (29), 91 (56).
C23H27NO6S [M+] requires *"/g 445.1559; observed 445.1557.
N-(Benzvloxvcarbonvl)-7a,8a-epoxv-9-azabicvclor4.2.l1nonane (259)Sodium hydride (60% dispersion, 97 mg, 2.43 mmol) and THF (1 ml) were stirred in
a 25ml two-necked flask under a nitrogen atmosphere at 0°C. A solution of (258) (451 mg, 1.01 mmol) in THF (4 ml) was injected, warmed to room temperature and stirred for 3 hr. The reaction was worked up using an identical procedure to that described for the exo-epoxide (235). Purification of the resultant cmde oil by flash chromatography, eluting with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C), afforded (259) (258 mg, 93%) as an oil which solidified on standing. An analytical sample was prepared by recrystallisation from toluene and petroleum ether (b.p. 40 - 60°C) to give a white solid (m.p. 49 - 50°C). The NMR shifts given in italics refer to signals common to both rotamers (in a 1:1 ratio).
ÔH (300 MHz, CDCI3): 1.43 (m, 2H), 1.71 - 2.11 (series of m, 6H), 3.87 (dd, J = 7.4,3.6 Hz, IH, HCO), 3.89 (dd, J = 7.4, 3.6 Hz, IH, HCO), 4.23 (m, 2H, a-N), 5.10 (s, 2H, CH2Ph), 7.33 (m, 5H).
ÔC (75 MHz, CDCI3): 25.0 & 25.2 (2 x CH2), 28.1 & 29.0 (2 x CH2), 55.6 & 55.8 (2 x CHO), 62.6 & 62.7 (2 x CHN), 66.6 (CH2Ph), 127.8, 127.9 & 128.4 (3 x aryl CH),
136.8 (aryl C), 152.7 (C=0).
172
Pmax (CH2CI2): 3040W, 2925m, 2860w, 1695s, 1440s, 1385w, 1370w, 1350w, 1325m, 1315m, 1300w, 1230w, 1200w, 1190w, 1155w, 1120m, 1095m, 1080w, 1030w, 1015w, 970w, 940w, 910m, 865w, 840w cm"^
% (%): 273 (M+, 28), 138 (12), 92 (8), 91 (100).
C16H19NO3 [M+] requires % 273.1365; observed 273.1370.
Found: C, 70.06; H, 7.02; N, 5.14%.C16H19NO3 requires: C, 70.31; H, 7.02; N, 5.12%.
N-Methvl-73,83-epoxv-9-azabicyclor4.2. llnonane (260)A solution of (235) (107 mg, 0.39 mmol) in dry THF (5 ml) was cooled with stiiTing to 0°C under a nitrogen atmosphere. A solution of DIB AH (IM in hexane, 2.16 ml,
2.16 mmol) was injected, warmed to room temperature and stirred for 3 hr. Excess hydride was destroyed by addition of the minimum quantity of water-saturated diethyl ether, the solution was dried oyer anhydrous sodium sulphate and then filtered through celite. The colourless solution was acidified with hydrogen chloride gas at 0°C and the solvent was distilled under reduced pressure. The residual white solid was repeatedly triturated with diethyl ether to remove benzyl alcohol yielding (260.HC1) (60 mg, 81%) as a hygroscopic white solid. The amine was volatile and NMR spectra of the free amine (260) were recorded by dissolving the salt in the minimum quantity of CDCI3, basifying with ammonia gas and filtering. The amine was stored as the hydochloride salt.
ÔH (300 MHz, CDCI3): 1.33 - 1.52 (series of m, 4H), 1.58 (m, 2H), 1.74 (m, 2H), 2.55 (s, 3H, CH3), 3.18 (dd, J = 7.0,1.6 Hz, 2H, a-N), 3.43 (s, 2H, HCO).
8c (75 MHz, CDCI3): 24.7 (2 x CH2), 30.8 (2 x CH2), 46.5 (CH3), 61.1 (2 x CHN),63.6 (2 X CHO).
Pmax (CDCI3): 3030w, 2930m, 2850w, 1445w, 1340w, 1305w, 1220w, 1165w, 1125w, 1080w, 1035w, 950w, 845w cm '\
% (%) [measured using (260.HC1)]: 153 (M+ - HCl, 41), 139 (23), 124 (22), 110
173
(100), 96 (85), 82 (48), 68 (36).
C9H15NO [M+ - HCl] requires % 153.1154; observed 153.1154.
73.8B-Epoxy-9-azabicvclor4.2. Hnonane (261)A solution of (235) (50 mg, 0.18 mmol) in dry methanol (4 ml) was hydrogenolysed using the procedure described for the conyersion of (194) into (197). Distillation of solyent under reduced pressure afforded (261) (22 mg, 86%) as a waxy solid.
6h (300 MHz, CDCI3): 1.32 - 1.83 (series of m, 8H), 2.25 (s, IH, NH), 3.34 (s, 2H, HCO), 3.42 (dd, J = 6.7, 1.7,2H, a-N).
ÔC (75 MHz, CDCI3); 24.7 (2 x CHg), 31.0 (2 x CHg), 56.4 (2 x CHO), 58.9 (2 x CHN).
Pmax (CH2CI2): 3350w, 3030w, 2930s, 2860m, 1445w, 1425w, 1395w, 1340w, 1310w, 1265brw, 1215w, 1200w, 1160w, 1135w, lllOw, 1040w, lOlOw, 950w, 930w, 860m, 840m, 825w, 815w, 795w cm'^.
% (%): 139 (M+, 54), 110 (61), 97 (27), 96 (100), 83 (43), 82 (50), 80 (20), 68 (28), 55 (43).
CgHi3NO [M+] requires ™/g 139.0997; observed 139.0998.
Homotropan-7a-ol (262)To a slurry of LAH (21 mg, 0.55 mmol) in dry THE (1 ml) was injected a solution of (259) (95 mg, 0.35 mmol) in THF (5 ml) which was then refluxed under nitrogen for 3 hr. Excess hydride was destroyed by the dropwise addition of the minimum quantity of water-saturated diethyl ether and the solution was then dried over anhydrous sodium sulphate. Filtration through celite and distillation of solvent under reduced pressure gave an oil which was purified by flash chromatography, eluting with 1:9 methanohdichloromethane saturated with ammonia gas, to afford (262) (38 mg, 70%) as a waxy solid.
Sh (300 MHz, CDCI3): 1.40 (ddd, J = 13.7, 6.4, 3.1 Hz, IH), 1.46 (m, IH), 1.66 (m, 5H), 1.90 (m, 2H), 2.50 (s, 3H, CH3), 2.63 (ddd, J = 13.7, 10.4, 9.5 Hz, IH), 3.18 (m.
174
IH, a-N), 3.24 (m, IH, a-N), 4.55 (dt, J = 10.4, 6.4, 6.4 Hz, IH, a-OH).
5c (75 MHz, CDCI3): 24.8 (2 x CH^), 26.7 & 33.6 (2 x CH^), 39.2 (CH3), 39.9 (CHg),61.1 & 65.3 (2 X CHN), 71.7 (CHOH).
Pmax (CH2CI2): 3620m, 3370bnn, 3040w, 2920s, 1465w, 1375w, 1260brw, 1180w, 1125w, 1090w, 1050m, lOlOw, 985w, 970w, 920w cm'^.
% (%): 155 (M+, 50), 149 (17), 138 (80), 112 (68), 111 (90), 110 (26), 100 (12), 97 (24), 96 (60), 91 (25), 83 (63), 82 (100).
C9H17NO [M+] requires "'/g 155.1310; observed 155.1312.
Norhomotropan-7a-ol (263)A solution of the epoxide (259) (38 mg, 0.14 mmol) in dry methanol (4 ml) was hydrogenolysed using the procedure described previously for the conversion of (194) into (197). Reduction of both carbamate and epoxide occured to afford (263) (15 mg, 76%) as a thick yellow oil.
Sh (300 MHz, CDCI3): 1.37 (ddd, J = 13.4, 7.7, 4.2 Hz, IH), 1.68 (m, 7H), 1.98 (m, IH), 2.43 (dt, J = 13.4, 10.0, J = 10.0 Hz, IH), 3.46 (m, IH, a-N), 3.57 (s, IH, HN), 3.62 (m, IH, a-N), 4.40 (dt, J = 10.0, 7.7, J « 7.7 Hz, IH, a-N).
Sc (75 MHz, CDCI3): 24.4, 24.5, 27.5, 35.0 & 38.1 (5 x CH2), 55.2 (CHN), 58.6 (CHN), 73.4 (CHOH).
Pmax (CH2CI2): 3610w, 3320brm, 3040w, 2920s, 2860m, 1445brw, 1335brw, 1255w, 1130w, 1070w cm'^
% (%): 141 (M+,32), 138 (17), 124 (39), 112 (18), 111 (20), 98 (37), 97 (100), 96 (34), 91 (33), 82 (55), 69 (43), 68 (47).
CgHigNO [M+] requires ™/g 141.1154; observed 141.1154.
N-Methvl-7a,8a-epoxv-9-azabicvclol4.2. llnonane (264)To a slurry of LAH (65 mg, 1.71 mmol) was injected a solution of (259) (140 mg.
175
0.51 mmol) in diethyl ether (7 ml) which was then stirred at -78°C for 3 hr. The reaction was then worked up as described previously for the exo-isomer (260); the amine was volatile and the solution was acidified with hydrogen chloride gas prior to
distillation of solvent under reduced pressure. Trituration with diethyl ether afforded (264.HCÎ) (52 mg, 54%) as a hygroscopic white solid. The free amine (264) was used to record NMR spectra which displayed a 10% impurity of the ring opened product (262).
ÔH (300 MHz, CDCI3): 1.45 - 2.01 (brm, 8H), 2.41 (s, 3H, CH3), 2.95 (brd, J = 4.1 Hz, 2H, a-N), 3.97 (d, J = 3.2 Hz, 2H, HCO).
8c (75 MHz, CDCI3): 25.7 (2 x CHg), 29.8 (2 x CHg), 46.7 (CH3), 65.3 (2 x CHN),65.8 (2 X HCO).
Pmax (CDCI3): 3 6 0 0 w , 30 2 0 W , 2 9 2 0 s , 2 8 6 0 w , 1 4 4 0 b rw , 1 3 8 0 w , 1 1 2 5 w , lllOw, 1 0 7 0 w , lOOOw, 9 0 5 w cm '\
% (%)[measured using (263.HC1)]: 153 (M+ - HCl, 45), 138 (16), 124 (31), 110 (100), 96 (69), 91 (55), 82 (42), 68 (29).
C9H15NO [M+ - HCl] requires "*/g 153.1154; observed 151.1154.
lB-Hvdroxv-2B,3B-epoxv-4B-f(benzvloxvcarbonyl)amino1cvcloheptane (266) and 13-Hydroxy-2a,3a-epoxy-4b-r(benzyloxycarbonyl)aminolcycloheptane (267)A solution of (227) (2.340 g, 8.97 mmol) in dichloromethane (160 ml) was epoxidised with MCPBA (50 - 60% purity, 3.38 g, 10.7 mmol) using the procedure described for the homologous ally lie alcohol (255). The crude solid obtained after distillation of solyent was purified by flash chromatography, eluting with diethyl ether, to afford
firstly (267) (741 mg, 30%) as a white solid which had m.p. 135 - 136°C after recrystallisation from ethanol.
ÔH (300 MHz, CD3COCD3): 1.42 (m, 2H), 1.57-1.81 (series of m, 6H), 3.01 (dd, J =5.8, 5.0 Hz, IH, HCO), 3.08 (dd, J = 6.6, 5.0 Hz, HCO, IH), 3.64 (m, IH), 3.79 (m, IH), 4.47 (brs, IH, exch), 5.07 (s, 2H, CH^Ph), 6.54 (brm, IH, NH), 7.37 (m, 5H).
8c (75 MHz, CDCI3): 18.9, 30.0 & 32.0 (3 x CHg), 51.0 (CHN), 56.7 & 57.7 (2 x
176
CHO), 66.8 (CHzPh), 70.6 (CHOH), 128.1 (2 x aryl CH), 128.5 (aryl CH), 136.4 (arylC), 155.8 (C=0).
Pmax (C H 2 C I2 ): 3 600W , 34 3 0 W , 3390brw, 2930m, 1 7 1 5 s , 1 5 1 0 m , 1 4 4 0 w , 1 4 2 0 b r w ,
1 3 1 5 w , 1220m, 1 1 3 0 w , 1 0 7 0 m , 1 0 2 5 m , 9 6 5 w , 9 3 5 w , 8 6 5 w , 8 5 0 w , 8 1 5 c m ' l
% (%): 277 (M+, 3), 124 (3), 122 (3), 108 (67), 107 (44), 106 (24), 105 (25), 92 (11), 91 (100), 79 (50).
Found; C, 65.16; H, 6.78; N, 4.89%.
C15H19NO4 requires: C, 64.96; H, 6.91; N, 5.05%.
Further elution with diethyl ether furnished (266) (1.572 g, 63%) as a white solid which had m.p. 151 - 152°C after reciystallisation from ethanol.
Sh (300 MHz, CDCI3): 1.01 (m, IH), 1.36 - 1.86 (series of m, 5H), 2.61 (brs, IH, exch), 3.22 (d, J = 5.0 Hz, IH, HCO), 3.29 (d, J = 5.0 Hz, IH, HCO), 3.96 (dd, J =11.4, 4.0 Hz, IH), 4.03 (m, IH), 5.10 (s, 2H, CH2Ph), 5.39 (d, J = 8.8 Hz, NH), 7.34 (m, 5H).
Sc (75 MHz, CDCI3): 22.0, 31.4 & 33.5 (3 x CH2), 51.8 (CHN), 58.3 & 60.2 (2 x CHO), 66.9 (CH2Ph), 71.5 (CHOH), 128.1, 128.2 & 128.5 (3 x aryl CH), 136.3 (aryl C), 155.7 (C=0).
Pmax (CH2CI2): 3600w, 3430w, 2940w, 2860w, 1720s, 1505m, 1445w, 1345w, 1305w, 1220m, lllOw, 1015m, 905w, 855w, 795w cm l
% (%): 277 (M+, 5), 183 (5), 108 (24), 107 (10), 92 (10), 91 (100).
Found: C, 65.29; H, 6.84; N, 4.97%.C15H19NO4 requires: C, 64.96; H, 6.91; N, 5.05%.
From ^H NMR signal integration of a crude sample. The ratio of (266): (267) was determined to be 62:38 in close agreement with the isolated yields.
177
Magnesium monoperperphthalate (MMPP) epoxidation of (227)Using the procedure of Gillard^^^ a solution of (227) (45 mg, 0.17 mmol) in ethanohwater (19:1, 5 ml) and MMPP (128 mg, 0.26 mmol) was stined for 3 hr at room temperature. The bulk of the solvent was then distilled under reduced pressure and the residual oil dissolved in dichloromethane (10 ml) and washed with water (5
ml). The organic layer was separated, dried over anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure. The crude oil was flash chromatographed to remove polar impurities, eluting with diethyl ether, affording(266) and (267) (34 mg, 72 %) in a 42:58 ratio (calculated from crude NMR integrations).
Vanadium-catalysed epoxidation of (227)The procedure of Teranishi^^^ was followed: a catalytic quantity of yanadium(III) acetylacetonate was added to a solution of (227) (65 mg, 0.25 mmol) in benzene (4 ml) and stirred for 5 min. An anhydrous solution of r-butyl hydroperoxide (3M in 2,2,4-trimethylpentane, 140 pi, 0.42 mmol) was injected and the pale green solution immediately turned red. Stirring was continued for 1 hr and TLC analysis indicated the majority of the mixture was still starting material. A further portion of peroxide (50 pi, 0.15 mmol) was added and stirred for a further 5 hr. The bulk of the solyent was distilled under reduced pressure and the residue was chromatographed with diethyl ether to remove polar impurities to give epoxides (28 mg, 40%) in a ratio of >95:<5 (266):(267), as calculated from NMR integrations.
23,3 b-Epoxv-43-r (benzyloxycarbonvl)aminolcvcloheptanone (269)A solution of (229) (61 mg, 0.24 mmol) in THF:H20 (4:1, 3 ml) was basified with sodium hydroxide solution (6M, 85 pi) at 0°C. An aqueous solution of hydrogen peroxide (30 weight %, 100 pi) was added, the solution warmed to room temperature and stirred for 4 hr. The bulk of the solvent was distilled and the residual oil was partitioned between diethyl ether (7 ml) and water (2 ml). The organic layer was washed with further water (2 ml), separated, dried over anhydrous magnesium sulphate and the solvent distilled under reduced pressure. The resulting crade solid was recrystallised from ethanol to afford (269) (45 mg, 69%) as a crystalline white solid (m.p. 147 - 148°C).
ÔH (300 MHz, CD3COCD3): 1.27 (m, IH), 1.83 (m, 3H), 2.21 (dd, J = 14.4, 4.3 Hz, IH), 2.62 (ddd, J = 14.4, 11.4, 2.9 Hz, IH), 3.37 (dd, J = 5.2, 0.9 Hz, IH, HCO, p-N),
178
3.47 (d, J = 5.2 Hz, IH, HCO), 3.97 (dd, J = 9.6, 4.3 Hz, IH), 5.10 (s, 2H, CHzPh),7.38 (m, 5H).
ÔC (75 MHz, CDCI3): 22.7, 30.6 & 40.7 (3 x CHg), 51.2 (CHN), 57.9 & 59.0 (2 x CHO), 67.0 (CH2PI1), 128.1, 128.2 & 128.5 (3 x aryl CH), 136.1 (aryl C), 155.5 (NC=0), 208.5 (C=0).
Pmax (CH2CI2): 3430m, 2940w, 2860w, 1715s, 1505s, 1450w, 1350w, 1305m, 1220s, 1115w, 1070w, 1015m, 925w, 960m cm'^
% (%): 275 (M+, 4), 108 (42), 107 (8), 92 (9), 91 (100).
Found; C, 65.68; H, 6.08; N, 5.26%.C15H17NO4 requires: C, 65.44; H, 6.23; N, 5.09%.
The ketone (269) could also be prepared in 87% yield by Jones oxidation^^ of the epoxy-alcohol (266) using an identical procedure to that described previously for the conversion of (132) into (133)
1B- F(p-T oluenesulphonvDoxvl -23.3 B-epoxv-4B- F (benzvloxvcarbonvDaminolcvclo- heptane (270)A solution of (266) (300 mg, 1.08 mmol) in THF (12 ml) was tosylated by sequential addition of n-butyllithium (2.5M in hexane, 0.52 ml, 1.30 mmol) and p-toluenesulphonyl chloride (269 mg, 1.41 mmol) in THF (4 ml), using the procedure described for tosylating the homologous epoxy-alcohol (244). The tosylate (270)
(460 mg, 99%) was obtained as a white solid after flash chromatography eluting with 3:2 diethyl ether:petroleum ether (b.p. 40 - 60°C). An analytical sample having m.p. 106 - 107°C was prepared by recrystallisation from toluene and petroleum ether (b.p. 60 - 80°C).
5h (300 MHz, CDCI3): 1.38 (m, 2H), 1.70 (m, 4H), 2.43 (s, 3H), 4.14 (d, J = 5.2 Hz, IH, HCO), 3.17 (d, J = 5.2 Hz, IH, HCO), 3.96 (m, IH), 4.74 (dd, J = 11.2, 4.4 Hz, IH), 5.07 (s, 2H, CHjPh), 5.36 (d, J = 8.7 Hz, IH, NH), 7.31 - 7.39 (m, including d, 7H), 7.90 (d, J = 8.5 Hz, 2H).
6c (75 MHz, CDCI3): 21.6 (CHg & CH3), 30.7 (CHj), 51.3 (CHN), 57.3 & 57.6 (2 x
179
CHO), 66.8 (CH^Ph), 81.7 (CHOSO ), 127.7, 127.8, 128.0, 128.1 & 129.8 (5 x arylCH), 133.7 (aryl CMe), 136.2 (aryl CCHg), 145.0 (aryl CSOg), 155.5 (C=0).
Pmax (C H 2 C I2 ): 34 3 0 W , 30 3 0 W , 2 9 3 0 w , 2 8 6 0 w , 1 7 1 5 s , 1 5 9 5 w , 1 5 0 0 s , 1 4 4 5 w , 1 3 6 0 s ,
1 3 0 5 m , 1 2 1 5 m , 1 1 8 5 s , 1 1 7 0 s , 1 1 0 5 w , 1 0 9 5 w , 1 0 2 0 w , 9 0 0 m , 8 6 5 m , 8 4 0 m , 8 1 0 m
cm’ .
% (%): 431 (M+, 2), 411 (2), 327 (2), 172 (3), 165 (3), 155 (4), 110 (9), 108 (13), 107(16), 92 (13), 91 (100).
C22H25NOQS [M+] requires 431.1403; obsei-ved 431.1406.
Found: C, 61.10; H, 5.73; N, 3.14%.
C22H25NO6S requires: C, 61.23; H, 5.84; N, 3.24%.
la-Chloro-2B.3B-epoxv-43-r(benzvloxvcarbonvl)amino1cycloheptane (271)A solution of (270) (670 mg, 1.55 mmol) in DMSO (12 ml) was chlorinated at 55°C with lithium chloride (410 mg, 9.76 mmol) using the procedure described for the prepaiation (246). The product (271) (407 mg, 85%) was obtained as a white solid after flash chromatography using 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C) and had m.p. 92 - 93 °C after recrystallising twice from ethanol.
§ H (300 MHz, CDCI3): 1.55 (m, 3H), 1.74 (m, IH), 1.91 (m, 2H), 3.30 (m, 2H, CHO),
4.21 (m, IH), 4.74 (m, IH), 5.11 (s, 2H, CH2ph), 5.17 (brd, J = 8.9 Hz, IH, NH), 7.34 (m, 5H).
8c (75 MHz, CDCI3): 19.3, 30.9 & 32.2 (3 x CH2), 51.2 (CHN), 56.8 (CHCl), 59.4 &61.1 (2 X CHO), 66.7 (CH2Ph), 128.1, 128.2 & 128.5 (3 x aryl CH), 136.3 (aryl C),155.5 (C=0).
Pmax (CH2CI2): 3430m, 3300brw, 3030w, 2940m, 1715s, 1495s, 1445w, 1395w, 1325w, 1305m, 1210s, 1175w, 1120w, 1105w, 1015m, 975w, 930w, 910w, 840w, 800w cm"\
% (%): 297 (M+, 1), 295 (M+, 3), 108 (42), 107 (17), 92 (9), 91 (100).
180
CisHigNOgCl [M+] requires % 295.0975; observed 295.0975.
Found; C, 60.61; H, 6.32; N, 4.60%.C15H18NO3CI requires: C, 60.91; H, 6.13; N, 4.74%.
N-(Benzvloxvcaibonvl)-6B,7B-epoxv-8-azabicvclor3.2.11octane (272)To a stirred slurry of sodium hydride (60% dispersion in mineral oil, 78 mg, 1.95 mmol) in diy THF:DME (8:1, 2 ml) was injected a solution of (271) (338 mg, 1.14
mmol) in THF:DME (8:1, 11 ml) at 0°C. The solution was stined at room temperature for 1 hr and then at 50°C for a further 1.5 hr. Excess hydride was
destroyed by the addition of water at -78°C and diethyl ether (25 ml) was added. The ethereal layer was washed with water ( 2 x 5 ml), brine (5 ml), separated and dried over anhydrous magnesium sulphate. After filtration and distillation of solvent under reduced pressure, the residual oil was purified by flash chromatography using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford (272) (256 mg, 87%) as an oil. On refrigeration the oil solidified and was reciystallised from toluene and petroleum ether (b.p. 60 - 80°C) to give a white solid (m.p. 67 - 68°C). The signals quoted in italics are common to both rotamers (in a 1:1 ratio).
ÔH (300 MHz, CDCI3): 1.46 -1.89 (series of m, 6H), 3.42 (d, J = 3.2 Hz, IH, HCO), 3.45 (d, J = 3.2 Hz, IH, HCO), 4.33 (brs, IH, a-N), 4.41 (brs, IH, a-N), 5.12 (s, 2H, CHgPh), 7.34 (m, 5H).
Sc (75 MHz, CDCI3): 16.8, 24.7 & 25.0 (3 x CHj), 51.5 & 51.9 (2 x CHO), 53.3 &53.8 (2 X CHN), 66.7 (CHgPh), 127.6, 127.7 & 128.3 (3 x aryl CH), 136.6 (aryl C),156.3 (C=0).
Pmax (CH2CI2): 3040m, 2950s, 2930s, 2880m, 2860m, 1700s, 1495m, 1425s, 1360s, 1330W, 1295s, 1235s, 1215m, 1100s, 1040s, 1030w, lOOOw, 980w, 940w, 910s, 900s cm'^
% (%): 259 (M+, 32), 172 (6), 152 (11), 125 (6), 92 (8), 91 (100), 65 (8).
Found: C, 69.40; H, 6.70; N, 5.32%.C15H17NO3 requires: C, 69.48; H, 6.61; N, 5.40%.
181
13-r(p-Toluenesulphonvl)oxv1-2a.3a-epoxv-43-r(benzvloxvcarbonyl~)amino1- cycloheptane (273)
A solution of (267) (739 mg, 2.67 mmol) in THF (18 ml) was tosylated by sequential
addition of n-butyllithium (2.5M in hexane, 1.17 ml, 2.94 mmol) and p-toluene- sulphonyl chloride (662 mg, 3.47 mmol) in THF (6 ml), using the procedure described
for tosylating the homologous epoxy-alcohol (244). The tosylate (273) (1.053 g, 92%) was obtained as a foam after flash chromatography eluting with 1:1 diethyl ethenpetroleum ether (b.p. 40 - 60°C).
ÔH (300 MHz, CDCI3): 2.56 (m, IH), 1.47 - 1.86 (series of m, 5H), 2.41 (m, 3H), 3.09 (m, 2H, HCO), 4.05 (m, IH), 4.88 (m, IH), 5.10 (s, 2H, CH^Ph, including brs, IH, NH), 7.28 (part of AA’BB’ 2H), 7.36 (m, 5H), 7.80 (part of AA'BB' 2H).
8c (75 MHz, CDCI3): 19.0 (CHg), 21.7 (CH3), 29.9 & 30.4 (2 x CHg), 51.0 (CRN),55.3 & 56.6 (2 x CHO), 66.9 (CHgPh), 81.6 (CHOSO^), 127.7, 128.2, 128.6 & 130.0
(4 X aryl CH), 133.5 (aryl CMe), 136.3 (aryl CCH^, 145.1 (aryl CSO2), 155.6 (C=0).
Umax (CH2CI2): 3440w, 3060V/, 2950w, 2870w, 1725s, 1600w, 1510s, 1465w, 1455w, 1365m, 1325w, 1230m, 1215m, 1190s, 1180s, 1090w, 1065w, 1030w, 1020w, 955m, 910m, 860w, 840w, 815w cm’^
% (%): 431 (M+, 9), 296 (5), 152 (8), 126 (6), 110 (19), 108 (30), 107 (34), 92 (16), 91 (100), 81 (10).
C22H25NO6S [M’’’] requires "'/z 431.1403; obseryed 431.1403.
la-Chloro-2cc,3a-epoxy-43-r(benzyloxycarbonyl)amino1cycloheptane (274)A solution of (273) (930 mg, 2.16 mmol) in DMSO was chlorinated with lithium chloride (552 mg, 13.1 mmol) at 55°C using the procedure described for the conyersion of (245) into (246). The product (274) (604 mg, 95%) was obtained as a white solid after flash chromatography, eluting with 2:3 diethyl ether:petroleum ether (b.p. 40 - 60°C), and had m.p. 75 - 76°C after recrystallisation from toluene and petroleum ether (b.p. 60 - 80°C).
6h (300 MHz, CD3COCD3): 1.36 (m, IH), 1.53 (m, 2H), 1.76 (m, 2H), 2.03 (m, IH), 3.19 (t, J = 4.6 Hz, HCO, p-N), 3.39 (d, J = 4.5 Hz, HCO, p-Cl), 4.55 (m, IH), 4.73
182
(dd, J = 11.9, 4.0 Hz, IH), 5.04 & 5.09 (ABq, J = 12.4 Hz, 2H, CH^Ph), 6.58 (brd, J =6.5 Hz, IH, NH), 7.36 (m, 5H).
5c (75 MHz, CDCI3): 20.0, 29.3 & 34.6 (3 x CHg), 49.6 (CHN), 57.0 (CHCl), 60.0 &61.1 (2 X CHO), 67.1 (CH2Ph), 128.3, 128.5 & 128.6 (3 x aryl CH), 136.0 (aryl C),156.0 (C=0).
'«max (CH2CI2): 3440m, 3030V/, 2940m, 2860w, 1740s, 1500s, 1325m, 1220s, 1080w, 1145w, 1120w, 1070m, 1040w, 1025w, lOOOw, 975w, 940w, 905w, 850w, 830m.
% (%): 297 (M+, 14), 295 (M+, 31), 216 (10), 108 (51), 107 (55), 108 (51), 109 (6), 110 (8), 92 (13), 91 (100).
C15H18NO3CI [M+] requires % 295.0975; observed 295.0973.
Found: C, 61.05; H, 6.25; N, 4.73%.C15H18NO3CI requires: C, 60.91 ; H, 6.13; N, 4.74%.
2(x,3a-Epoxv-43-r(benzvloxvcarbonvl)aminolcvcloheptanone (276)
A solution of (267) (114 mg, 0.41 mmol) was oxidised with Jones reagent^^ using the procedure described for the conversion of (132) into (133). The ketone (276) (108 mg, 95%) was obtained as a colourless oil and was pure enough for the next reaction without purification.
ÔH (300 MHz, CDCI3): 1.24 (m, IH), 1.63 (m, IH), 1.76 - 2.03 (m, 2H), 2.28 (m, IH), 2.61 (m, IH), 3.38 (dd, J = 4.8, 1.2 Hz, IH, HCO), 3.42 (t, J = 4.8 Hz, IH, HCO, P-N), 4.79 (m, IH), 5.05 & 5.10 (ABq, J = 12.1 Hz, 2H, CHzPh), 5.32 (d, J = 8.7 Hz, IH, NH), 7.31 (m, 5H).
5c (75 MHz, CDCI3): 16.7, 28.2 & 40.2 (3 x CHg), 48.8 (CHN), 55.6 & 58.4 (2 x
CHO), 67.0 (CHjPh), 128.1, 128.2 & 128.5 (3 x aryl CH), 136.0 (aryl C), 155.7 (NC=0), 209.3 (C=0).
Pmax (CH2CI2): 3430m, 2950m, 2870w, 1715s, 1505s, 1450w, 1420w, 1400w, 1325m, 1215s, 1145w, 1120w, 1060m, 1025w, 985m, 935w, 905w, 880w, 860w, 825w cm'^.
183
% (%): 275 (M+, 3), 169 (3), 108 (26), 107 (12), 91 (100).
C15H17NO4 [M" ] requires "'/g 275.1158; observed 275.1157.
la-Hvdroxv-2a,3«-epoxv-4B-r(benzvloxvcarbonvl)amino1cvcloheptane (277)A solution of (276) (234 mg, 0.85 mmol) in THF (14 ml) was gently refluxed with sodium borohydride (79 mg, 2.14 mmol) for 1 hr before destroying excess hydride
with saturated ammonium chloride solution (1 ml). Diethyl ether (25 ml) was added,
the solution transfered to a separating funnel, and washed with water ( 2 x 5 ml) and brine (5 ml). The organic layer was dried over anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure to afford a mixture of epimers (223 mg, 95%) as an oil. The ^H NMR spectrum was free of impurities and the ratio of (267):(277) was calculated to be 62:38 respectively from signal integrations. The epimers were separated by flash chromatography, eluting with 4:1
diethyl ether:petroleum ether (b.p. 40 - 60°C), to afford firstly the cis-1,4 alcohol(267) which had identical NMR spectra to a sample prepared by epoxidation of (227). Further elution afforded the trans-1,4 alcohol (277) as a white solid which had m.p. 128 - 129°C after recrystallisation from toluene.
ÔH (300 MHz, CDCI3): 1.17 (m, IH), 1.67 (m, 5H), 2.80 (brs, IH, exch), 3.12 (m, IH, HCO, p-N), 3.17 (dd, J = 4.7, 1.6, IH, HCO, p-OH), 4.08 (m, IH), 4.34 (m, IH), 5.08 (brs, 3H, CHzPh and NH), 7.34 (m, 5H).
6c (75 MHz, CDCI3): 18.5, 30.5 & 32.0 (3 x CHj), 50.5 (CHN), 56.3 & 59.3 (2 x CHO), 67.0 (CHzPh), 69.2 (CHOH), 128.1, 128.2 & 128.5 (3 x aryl CH), 136.2 (aryl C), 155.7 (C=0).
«max (CH2CI2): 3590w, 3440w, 2930w, 2860w, 1720s, 1505s, 1450w, 1395w, 1325w, 1225m, 1210m, 1070w, 1025w, lOOOw, 935w, 905m, 835w, 805w cm"\
% (%): 277 (M+, 53), 186 (32), 168 (11), 124 (15), 107 (16), 91 (100).
C15H19NO4 [M+] requires ™/g 277.1314; observed 277.1315.
Found: C, 64.66; H, 6.95; N, 4.91%.C15H19NO4 requires: C, 64.96; H, 6.91; N, 5.05%.
184
L-Selectride Reduction of (276)
A solution of (276) (640 mg, 2.33 mmol) in THF (25 ml) was cooled with stirring to
-78°C. L-Selectride (IM solution in THF, 2.79 ml, 0.28 mmol) was slowly injected and stirring was continued at reduced temperature for 1 hr. The solution was quenched with water (200 pi) and warmed to room temperature. The bulk of the solvent was distilled under reduced pressure and the residual oil was partitioned between diethyl ether (55 ml) and sodium hydroxide solution (IM, 10 ml) and then washed with water (7 ml) and brine (7 ml). The organic layer was dried over
anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure. The crude oil was purified by flash chromatography, eluting with 4:1 diethyl ether:petroleum ether (b.p. 40 - 60°C). The first fraction to be eluted was the cw-1,4 alcohol (267) (262 mg, 41%). Further elution afforded the trans-\,A alcohol (277) (313 mg, 48%).
la-r(p-Toluenesulphonvl)oxvl-2a.3q-epoxv-43-r(benzvloxvcarbonvl)amino1- cvcloheptane (278)A solution of (277) (220 mg, 0.79 mmol) in THF (8 ml) was tosylated by sequential addition of n-butyllithium (2.5M in hexane, 0.41 ml, 1.03 mmol) and p-toluenesulphonyl chloride (197 mg, 1.03 mmol) in THF (2 ml), using the procedure described for tosylating the epoxy-alcohol (244). The tosylate (278) (303 mg, 89%) was obtained as a white solid after flash chromatography eluting with 3:2 diethyl ether:petroleum ether (b.p. 40 - 60°C). An analytical sample was prepared by recrystallising from toluene and petroleum ether (b.p. 60- 80°C) which had m.p. 139 - 140°C.
ÔH (300 MHz, CDCI3): 1.03 (m, IH), 1.46 (m, 2H), 1.76 (m, 3H), 2.42 (s, 3H), 3.14 (bit, J = 4.5 Hz, HCO, p-N), 3.25 (d, J = 4.5 Hz, HCO, p-OSO^Ar), 4.56 (brm, IH), 4.98 (brm, IH), 5.08 (brs, 2H, CHzPh), 5.15 (brd, J = 8.6 Hz, IH, NH), 7.33 (m, 7H), 7.79 (part of AA’BB’, 2H).
Sc (75 MHz, CDCI3): 17.0 (CHj), 21.7 (CH3), 29.0 & 30.4 (2 x CHg), 49.3 (CHN),54.2 & 58.5 (2 x CHO), 67.1 (CHgPh), 81.4 (CHOSOg), 127.6, 128.2, 128.5 & 129.9 (4 X aryl CH), 133.9 (aryl CMe), 136.1 (aryl CCHj), 144.9 (aryl CSOg), 155.7 (C=0).
«max (CH2CI2): 3430w, 2930w, 2870w, 1720s, 1600w, 1500s, 1445m, 1400w, 1360s,
185
1330s, 1215s, 1190s, 1185s, 1220w, 1095m, 1085m, 1005w, 950s, 895s, 850m, 835m, 815m cm‘^
% (%): 431 (M+, 21), 307 (8), 259 (6), 107 (54), 91 (100).
C22H25NO6S [M+] requires "'/g 431.1403; observed 431.1406.
Found: C, 61.27; H, 5.81; N, 3.13%.C22H25NOgS requires: C, 61.23; H, 5.84; N, 3.25%.
N-(Benzvloxvcarbonvl)-6a,7a-epoxv-8-azabicvclor3.2. lloctane (275)A 25 ml oven dried two-necked flask was charged with sodium hydride (60% dispersion in mineral oil, 30 mg, 0.75 mmol) and slurried with dry THF:DME (5:1, 1 ml). The system was fitted with a condenser and septum cap and repeatedly purged and evacuated with nitrogen. A solution of (278) (55 mg, 0.13 mmol) in THF:DME (5:1, 4 ml) was injected and the solution was stilted at room temperature for 1 hr and then at 40°C for a further 1 hr. After cooling to -78°C the reaction was quenched by the addition of water (200 pi). Diethyl ether (10 ml) was added and washed with water ( 2 x 3 ml) and brine (3 ml) and then dried over anhydrous magnesium sulphate.
Distillation of solvent under reduced pressure gave an oil which was purified by flash chromatography, eluting with 1:4 diethyl ether:petroleum ether (b.p. 40 - 60°C), to furnish (275) (20 mg, 61%) as a pale yellow oil. Restricted rotation about the N-CO bond caused some NMR signals to be broadened (italics) as the spectra were recorded close to the coalescence temperature.
ÔH (300 MHz, CDCI3): 1.26 (m, 2H), 1.73 (m, 2H), 1.98 (m, 2H), 3.95 (brd, J = 2.8 Hz, 2H, HCO), 4.24 (brm, 2H, a-N), 5.12 (s, 2H, CH2ph), 7.34 (m, 5H).
ÔC (75 MHz, CDCI3): 16J , 22.7 & 23.4 (3 x CH2), 53.3 (2 x CHN), 62.9 (2 x CHO),66.8 (CH2Ph), 127.8,128.0 & 128.5 (3 x aryl CH), 136.7 (aryl C), 152,9 (C=0).
«max (CH2CI2): 2930w, 1690s, 1495w, 1410m, 1360w, 1310m, 1265brw, 1220w, 1100m, 1065w, 1040w, 955w, 900w, 855w cm '\
% (%): 259 (M+, 3), 172 (4), 124 (9), 92 (8), 91 (100).
186
C15H17NO3 [M+] requires ""/g 259.1208; observed 259.1206.
1 B-(Mesvloxv)-2a.3a-epoxv-4B-r(benzvloxvcarbonvl)aminolcvclooctane (279)
Triethylamine (293 pi, 2.11 mmol) and (255) (409 mg, 1.40 mmol) in
dichloromethane (25 ml) were stirred at 0°C under a nitrogen atmosphere. Methanesulphonyl chloride (120 pi, 1.78 mmol) was injected and the solution was stirred at room temperature for 1 hr. The solution was transferred to a separating funnel and washed with hydrochloric acid (IM, 5 ml), saturated sodium bicaibonate
solution (5 ml) and water (5 ml). The organic layer was separated, dried over anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure to afford (279) (505 mg, 97%) as a foam which was determined to be pure from NMR spectroscopy. The foam solidified on refrigeration and an analytical sample was prepared by recrystallising from toluene and petroleum ether (b.p. 60 - 80°C) to give a white solid (m.p. 138 - 140°C).
ÔH (300 MHz, CDCI3): 1.38 - 1.69 (brm, 5H), 1.83 (m, 2H), 2.04 (m, IH), 3.07 (brs,
4H, MeSOg & HCO), 3.16 (dd, J = 8.3, 4.6 Hz, IH, HCO), 3.49 (m, IH, a-N), 4.48 (m, IH, a-OSOzMe), 5.05 & 5.11 (ABq, J = 11.3 Hz, 2H, CHgPh), 5.46 (d, J= 7.8 Hz, IH, NH), 7.33 (m, 5H).
Sc (75 MHz, CDCI3): 22.8, 23.9, 33.3 & 33.7 (4 x CHg), 38.3 (CH3), 51.4 (CHN),56.7 & 57.7 (2 x CHO), 66.8 (CHjPh), 83.6 (CHOSOg), 128.1 (2 x aryl CH), 128.5 (aryl CH), 136.3 (aryl C), 155.6 (C=0).
« m a x (C H 2 C I2 ): 3 4 3 0 m , 3 0 3 0 W , 2 9 4 0 m , 2 8 7 0 w , 1 7 2 0 s , 1 5 1 0 s , 1 4 5 0 m , 1355s, 1 3 0 0 w ,
1 2 2 5 s , 1 1 7 5 s , 1125W, 1 0 9 0 w , 1 0 2 5 m , 9 8 0 m , 9 4 0 s , 9 0 0 w , 8 8 0 w , 8 6 0 m cm l
% (%): 369 (M+, 14), 149 (6), 124 (5), 108 (25), 107 (30), 91 (100), 79 (12).
Found: C, 55.24; H, 6.32; N, 3.76%.C17H23NO6S requires: C, 55.28; H, 6.27; N, 3.79%.
N-Methvl-6B,7B-epoxv-8-azabicvcloF3.2. lloctane (280)
A solution of (272) (70 mg, 0.27 mmol) in diethyl ether (5 ml) was cooled with stirring to -78°C. A solution of DIB AH (IM in hexane, 0.95 ml, 0.95 mmol) was injected and stirred at -78°C for 45 min. A further portion of DIB AH (0.40 ml, 0.40
187
mmol) was added and stirred for a further 2 hr. The reaction was quenched with water (60 pi), warmed to room temperature and dried with anhydrous sodium sulphate. The solution was filtered through celite and acidified with gaseous hydrogen chloride at 0°C. The solvent was distilled under reduced pressure and the residual oil was repeatedly triturated (to remove benzyl alcohol) with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford (280.HC1) (27 mg, 57%) as a hygroscopic white solid. All spectra below were recorded on the free amine by dissolving (280.HC1) in the minimum quantity of CDCI3, basifying with ammonia gas and filtering.
6h (300 MHz, CDCI3): 1.45 (m, 2H). 1.63 - 1.78 (m, 2H), 2.52 (s, 3H), 3.12 (m, 2H, a-N), 3.56 (s, 2H, HCO).
ÔC (75 MHz, CDCI3): 16.7 (CHj), 24.3 (2 x CHg), 41.5 (CH3), 55.7 (2 x CHN), 58.8 (2 X CHO).
«max (CDCI3): 2930s, 1475w, 1440w, 1385w, 1330w, 1280w, 1080w, 1040w, 1015w, 905s, 850m cm*^
% (%): 139 (M+, 91), 110 (51), 96 (100), 94 (16), 91 (25), 82 (28), 70 (22).
CgHi3N0 [M+3 requires 139.0997; observed 139.0999.
6B,7B-Epoxv-8-azabicvclol3.2. lloctane hvdrochloride (281.HC1)A solution of (272) (66 mg, 0.25 mmol) in absolute ethanol (8 ml) was hydrogenolysed for 3 hr at 1 atmosphere using the procedure described for the conversion of (194) into (197). The solution was acidified with hydrogen chloride gas prior to distillation of solvent to afford (281.HC1) (39 mg, 95%) as a hygroscopic white solid. All spectral data quoted below was recorded on the hydrochloride salt, rather than the free amine.
Sh (300 MHz, CD3OD): 1.61 (m, IH), 1.82 - 2.11 (series of m, 5H), 3.94 (s, 2H, CHO), 4.01 (brs, 2H, a-N).
8c (75 MHz, CDCI3): 16.6 (CHj), 23.7 (2 x CHj), 52.2 (2 x CHN), 55.1 (2 x CHO).
188
% (%): 125 (M+ - HCI, 46), 108 (16), 96 (100), 82 (56), 69 (33), 56 (14).
C7H11NO [M^ - HCI] requires ™/g 125.0841; observed 125.0840.
Tropan-6B-ol (282)A solution of (272) (79 mg, 0.31 mmol) in diethyl ether (6 ml) was cooled with stirring to 0°C. A solution of DIB AH (IM in hexane, 1.68 ml, 1.68 mmol) was injected and stirred under a nitrogen atmosphere for 3 hr. The solution was quenched with the minimum quantity of water-saturated diethyl ether and the salts that precipitated were filtered off through celite. The colourless solution was acidified with gaseous hydrogen chloride gas and the solvent was distilled under reduced pressure. The residual oil was triturated with 1:1 diethyl ether:petroleum ether (b.p.
40 - 60°C) to afford (282.HCI) (38 mg, 70%) as a gum. Dissolving the salt in the minimum quantity of CDCI3, basifying with ammonia gas and filtering gave a sample of the free amine (282) for NMR and IR spectroscopy.
ôjj (300 MHz, CDCI3): 1.12 (m, IH), 1.24 - 1.54 (series of m, 3H), 1.66 - 1.87 (series of m, 2H), 1.98 (ddd, J = 13.6, 6.9, 3.1 Hz, IH), 2.09 (dd, J = 13.6, 7.2 Hz, IH), 2.54
(s, 3H), 2.96 (brs, IH, a-N), 3.29 (m, IH, a-N), 4.26 (dd, J = 7.2, 3.1 Hz, IH, a-N).
ÔC (75 MHz, CDCI3): 17.7, 24.4 & 25.7 (3 x CHj), 37.6 (CH3), 40.0 (CHg), 60.8 &66 .6 (2 X CHN), 75.7 (CHOH).
«max (CDCI3): 3300vbrm, 2930m, 1640w, 1440w, 1340w, 1255w, 1230w, 1220w, 1130w, 1030s, 910brmcm \
% (%): 141 (M+ - HCI, 34), 117 (14), 97 (100), 96 (39), 91 (52), 82 (30).
CgHjsNQ [M" - HCI] requires ’"/g 141.1154; observed 141.1155.
Tropan-6a-ol (283)A solution of (275) (48 mg, 0.18 mmol) in diethyl ether (5 ml) was refluxed with LAH (26 mg, 0.70 mmol) under a nitrogen atmosphere for 1 hr. Excess hydride was destroyed by the addition of water saturated diethyl ether and the solution was dried with anhydrous magnesium sulphate. Filtration through celite and distillation of solvent under reduced pressure gave a crude oil which was purified by flash
189
chromatography, using 7:3 diethyl ether:petroleum ether (b.p. 40 - 60°C) to elute benzyl alcohol and 1:1:3 triethylamine:methanol:ethyl acetate to elute (283) (18 mg, 69%) as a yellow oil.
ÔH (250 MHz, CDCI3): 1.28 (m, IH), 1.43 - 2.06 (series of m, 6H), 2.47 (s, 3H), 2.59 (ddd, J = 13.5, 10.6,7.5 Hz, IH), 3.06 (m, IH, a-N), 3.13 (m, IH, a-N), 3.87 (brs, IH, exch), 4.66 (ddd, J = 10.6, 6.0,4.1 Hz, IH, a-OH).
6c (63 MHz, CDCI3): 17.1, 21.7, 26.7 & 37.1 (4 x CH^), 37.7 (CH3), 59.8 & 63.1 (2 x CHN), 71.9 (CHOH).
Pinax (CH2CI2): 3320brs, 3050s, 2990s, 2940s, 1440s, 1420s, 1375w, 1355w, 1265s, 1150brw, 1120w, 1105w, 1090w, 1070m, 1005m, 990w, 970w, 960w, 895s, 850 cm"\
% (%): 141 (M+, 13), 124 (3), 112 (4), 97 (100), 82 (57), 68 (10).
CgHjsNG [MT requires ™/g 141.1154; observed 141.1154.
Cvclohepta-3,5-dienol (302)Using a modified version of the procedure of Schiess and Wisson,^^^ peroxyacetic acid (32 weight %, 117.0 g, 0.49 mol) was dripped into a stirred solution of cycloheptatriene (300) (32.4 g, 0.35 mol) in dichloromethane (300 ml) containing anhydrous sodium carbonate (85.0 g, 0.80 mol) over 30 min at 0°C. Stimng was continued for a further 3 hr at 0°C and the solution was then filtered through celite with thorough washing of the filter cake with dichloromethane. The solution was transferred to a separating funnel and washed with saturated sodium bicarbonate
solution (2 X 60 ml) and brine (50 ml). The organic layer was separated, dried over anhydrous sodium sulphate, filtered, and the solvent distilled at atmospheric pressure. The crude epoxide (301) was then dissolved in diethyl ether (100 ml) and dripped into
a slurry of LAH (5.70g, 0.15 mol) in diethyl ether (240 ml) at 0°C. After complete addition (30 min) the mixture was stirred for a further 1 hr and an aliquot was removed for NMR analysis which indicated complete reaction. Excess hydride was destroyed by the careful addition of sodium hydroxide solution (2 M) and the mixture was dried over anhydrous sodium sulphate. Filtration through celite and distillation of solvent under reduced pressure gave a crude yellow oil. Vacuum distillation of the oil at 20 mbai', 110°C removed volatile impurities. Further distillation at 5 mbar.
190
110°C furnished (302) (9.10 g, 23%) as a colourless oil.
ÔH (90 MHz, CDCI3): 2.50 (~t, J = 4.5 Hz, 4H), 4.10(m, IH, a-OH), 5.40 - 5.60 (series of m, 4H).
N-(Benzvloxvcarbonvl)-3a-hvdroxy-6-aza-7oxabicvclo-r3.2.21non-8-ene (303) and
N-(benzvloxvcarbonvl)-3B-hvdroxv-6-aza-7oxabicyclo-r3.2.21non-8-ene (304)A solution of benzyl-N-hydroxycarbamate (5.91 g, 0.031 mol) in dichloromethane (12 ml) was added to a solution of cycloocta-3,5-dieneol (302) (3.42 g, 0.031 mol) and tétraméthylammonium periodate (8.24 g, 0.031 mol) in dichloromethane (45 ml). An identical reaction and work-up procedure to that described for the preparation of the
unsubstituted cycloadduct (221) was used. An inseparable mixture of stereoisomers (303);(304) (6.21 g, 73%) was obtained in a 35:65 ratio (calculated from ^H NMR signal integrations) as a pale yellow oil after flash chromatography using 9:1 diethyl ether:petroleum ether (b.p. 40 - 60°C). The full assignment of NMR signals was
possible after the preparation of an enriched sample of the 3a-hydroxy isomer (303), described later. Signals common to both isomers are quoted in italics.
5h (300 MHz, CDCI3), 3a-Hydroxy isomer (303): 1.99 (m, IH), 2.04 (m, IH), 2.44 (ddd, J « 14 Hz, J = 5.5, 4.1 Hz, IH). 2.47 (ddd, J = 14 Hz, J = 5.5, 4.5 Hz, IH), 2.55 (brs, exch, OH), 4.24 (~quin, J = 5.5 Hz, IH, a-OSi), 4.74 (m, IH, a-N), 4.90 (m, IH, a-0), 5.17 (s, 2H, CHiPh), 6.39 (ddd, J = 9.1, 6.4, 1.3 Hz, IH, HC=), 6.50 (ddd, J =9.1, 6.4, 1.1 Hz, IH, HC=), 7.33 (m, 5H).
6c (75 MHz, CDCI3): 38.4 & 40.8 (2 x CHg), 51.6 (CHN), 66.9 (CHOH), 67.85 (CH^Ph), 72.0 (CHO), 128.0, 128.2, & 128.4 (3 x aryl CH), 129.7 & 132.3 (2 x CH=),135.9 (aryl C), 156.1 (C=0).
6h (300 MHz, CDCI3), 3p-Hydroxy isomer (304): 1.79 - 1.98 (m, 2H), 2.24 (m, 2H), 2.55 (brs, exch, OH), 3.67 (~tt, J = 6.2, 4.4 Hz, IH, a-OSi), 4.74 (m, IH, a-N), 4.90 (m, IH, a-O), 5.15 (s, 2H, CH^Ph), 6.17 (ddd, J = 9.1, 6.2, 1.2 Hz, IH, HC=), 6.32 (ddd, J = 9.1,6.8,0.6 Hz, IH, HC=), 7.33 (m, 5H).
6c (75 MHz, CDCI3): 36.8 & 39.6 (2 x CH2), 51.2 (CHN), 65.6 (CHOH), 67.77 (CHgPh), 72.4 (CHO), 128.0, 128.2, & 128.4 (3 x aryl CH), 129.7 & 132.0 (2 x CH=),135.9 (aryl C), 156.4 (C=0).
191
Vmax (CH2CI2), Mixture of (303) and (304); 3610w, 3490brw, 3040w, 2960s, 2930s, 2880w, 1705s, 1500w, 1455m, 1395m, 1355m, 1270brs, 1215w, 1175w, 1075s, 1045s, 1030w, 925w, 860w, 820w cm"^
™/z (%), mixture of (303) and (304): 275 (M+, 2), (231 (2), 92 (7), 91 (100).
C15H17NO4 [M+] requires ”*/z 275.1158; observed % 275.1158.
N-(Benzvloxvcaibonyl)-6-aza-7oxabicvclor3.2.21non-8-en-3-one (305)A solution of (303) and (304) (6.26 g, 0.023 mol) in acetone (110 ml) was oxidised with Jones reagent^® using the procedure described for the conversion of (132) into (133). The ketone (305) (5.68 g, 90%) was isolated as a pale yellow solid and was pure enough for the subsequent reaction without purification. An analytical sample
was prepared by recrystallising from toluene and petroleum ether (b.p. 60 - 80°C) to give a crystalline white solid (m.p. 82 - 83°C).
ÔH (300 MHz, CDCI3): 2.69 (m, 2H), 3.05 (dd, J « 13 Hz, J = 2.8 Hz, IH), 3.12 (dd, J = 13 Hz, J = 3.9 Hz, IH), 4.90 (m, IH, a-N), 4.97 (m, IH, a-O), 5.16 & 5.21 (ABq, J = 12.4 Hz, 2H, CH2Ph), 6.40 (ddd, J = 9.2, 6.2, 1.3 Hz, IH, HC=), 6.57 (ddd, J = 9.2,6.8, 0.9 Hz, IH, HC=), 7.34 (m, 5H).
8c (75 MHz, CDCI3): 47.1 & 49.8 (2 x CH2), 50.6 (CHN), 66.1 (CH2Ph), 70.6 (CHO),128.1, 128.3, & 128.5 (3 x aryl CH), 131.4 & 132.7 (2 x CH=), 135.6 (aryl C), 156.6 (NC=0), 206.0 (C=0).
Vmax (CH2CI2): 3040w, 2950W, 1710s, 1500w, 1450w, 1395m, 1375m, 1355m, 1330m, 1270brs, 1215m, 1150w, 1095m, 1070s, 1040m, lOOOw, 980w, 910m, 815w cm‘ .
% (%): 273 (M+, 2), 229 (9), 92 (11), 91 (100).
C15H15NO4 [M+] requires ™/z 273.1001; observed 273.1002.
Found: C, 66.00; H, 5.61; N, 5.15%.C15H15NO4 requires: C, 65.92; H, 5.53; N, 5.13%.
192
N-(Benzvloxvcai-bonvl)-3a-hvdroxv-6-aza-7oxabicvclor3.2.21non-8-ene (303)A solution of (305) (5.58 g, 0.020 mol) in THF (110 ml) was cooled with stirring to -78°C under a nitrogen atmosphere. A solution of L-Selectride (IM in THF, 21.5 ml, 0.022 mol), was slowly dripped in (down the side of the flask) over 30 min. The solution was stirred at -78°C for 2 hr, quenched with water (1 ml) and warmed to 0°C. Sodium hydroxide solution (IM, 20 ml) and aqueous hydrogen peroxide (30 weight %, 5 ml) were added and stirred for 10 min. The mixture was transferred to a separating funnel and diethyl ether (100 ml) was added. The aqueous layer was
separated off and the ethereal layer was washed with water (20 ml) and brine (10 ml). After drying over anhydrous magnesium sulphate and filtration, the bulk of the
solvent was distilled under reduced pressure. The residual oil was purified by flash chromatography, eluting with diethyl ether, to give (303):(304) (4.78 g, 85%) in a ratio of 85:15 as determined from ^H NMR integrations. The two isomers were inseperable at this stage of the synthesis.
N-Benzvloxvcarbonvl-3oc-r(r-butvldimethvlsilvBoxvl-6-aza-7oxabicvclol3.2.21- non-8-ene (306)The 85:15 mixture of (303):(304) (4.18 g, 0.015 mol) was dissolved in DMF (15 ml) and cooled to 0°C. Imidazole (1.67 g, 0.024 mol) and TBDMSCl (3.02 g, 0.020 mol) were added and stirred for 1.5 hr. Water (50 ml) was added, the mixture transferred
to a separating funnel and extracted with diethyl ether (3 x 70 ml). The combined ethereal layers were washed with water (15 ml) and brine (10 ml). After drying over
anhydrous magnesium sulphate and filtration, the bulk of the solvent was distilled under reduced pressure. The residual oil was purified by flash chromatography using 1:9 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford (306) (5.44 g, 91%) as a thin yellow oil. Partial separation of (306) from the 3p-silylether (308) was possible
by chromatography to give an improved ratio of 90:10 (from ^H NMR integrations) but minor traces of (308) were still visible in the NMR spectra; this data is quoted in full later.
ÔH (300 MHz, CDCI3): 0.05 [s, 6H, ( ^ 3)281], 0.88 [s, 9H, (CH3)3CSi], 1.76 (m, 2H),2.42 (m, 2H), 4.40 (~tt, J = 7.7, 5.9 Hz, IH, a-OSi), 4.77 (brt, J = 4 Hz, IH, a-N),4.92 (brt, J « 5 Hz, IH, a-O), 5.19 & 5.24 (ABq, J = 12.3 Hz, 2H, CH2Ph), 6.35 (ddd, J = 9.1, 6.4,1.3 Hz, IH), 6.45 (ddd, J = 9.1,7.1,1.0 Hz, IH), 7.31 - 7.40 (m, 5H).
193
Sc (75 MHz, CDCI3): -4.9 [ ( ^ 3)28!], 17.7 [(CH3)3CSi], 25.6 [(CHa))^!], 36.5 &40.7 (2 X CH2), 51.0 (CHN), 67.6 (CHOSi), 67.8 (CH2?h), 72.4 (CHO), 127.9, 128.0 & 128.3 (3 X aryl CH), 131.5 & 131.6 (2 x CH=), 135.9 (aryl C), 156.2 (C=0).
Vmax (CH2CI2): 2930m, 2850m, 1690s, 1555w, 1380w, 1350m, 1290w, 1260w, 1135w, 1080s, 1030w, 1005w, 925w, 905m, 835m, 740vbnn.
"i/z (%): 389 (M+, 3), 332 (5), 288 (8), 273 (16), 167 (13), 147 (8), 91 (100).
C2iH3iN04Si [M+] requires "'/z 389.2022; observed 389.2022.
6-r(^Butvldimethvlsilvl)oxvlcvclohepta-1.3-diene (307)A solution of (302) (1.506 g) in dry DMF (24 ml) was stirred at 0°C under a nitrogen atmosphere. Imidazole (1.489 g, 21.9 mmol) and TBDMSCl (2.687 g, 17.8 mmol) were added and stirred for 2.5 hr. The reaction mixture was poured into water (30 ml)
and repeatedly extracted with diethyl ether (2 x 75 ml, 1 x 50 ml). The combined ethereal layers were washed with water (20 ml), brine (10 ml) and then dried over anhydrous magnesium sulphate. Filtration and distillation of solvent under reduced pressure gave a crude oil which was purified by flash chromatography, eluting with petroleum ether (b.p. 40 - 60 °C) to afford (307) (2.847 g, 93%) as a colourless oil.
Sh (300 MHz, CDCI3): 0.12 [s, 6H, (€ « 3)28]], 0.95 [s, 9H, (CH3)3CSi], 2.51 (m, 4H), 4.11 (tt, J = 8.0,4.8 Hz, IH, a-OSi), 5.71 (m, 2H), 5.82 (m, 2H).
Sc (75 MHz, CDCI3): -4.8 [(CH3)2Si], 18.1 [(€«3)3081], 25.9 [(CH3)3CSi], 40.9 (2 x CH2), 71.4 (CHOSi), 126.0 (2 x CH), 127.9 (2 x CH).
Vmax (CH2CI2): 3010W, 2930s, 2895s, 2850s, 1465m, 1380m, 1360m, 1240w, 1075brs, 1030s, 1005m, 965w, 940w, 910s, 875w, 835s, 720vbrm cm \
N-Benzvloxvcarbonvl-3B-([r-butvldimethvlsilvl)oxvl-6-aza-7oxabicvclo[3.2.21- non-8-ene (308)Tétraméthylammonium periodate (4.04 g, 0.015 mol) and (307) (2.85 g, 0.013 mol) in dichloromethane (65 ml) were stirred at -78°C. A solution of benzyl-N-hydroxy- carbamate (2.55 g, 0.015 mol) in dichloromethane (20 ml) was dripped in over 10 min and the solution was then warmed to ambient temperature and stirred for 1.5 hr. The
194
reaction was worked up using an identical procedure to that described for the preparation of the unsubstituted cycloadduct (221). Purification of the crude oil by flash chromatography, eluting with 1:4 diethyl etherrpetroleum ether (b.p. 40 - 60°C), afforded (308):(306) (4.18 g, 85%) in an 80:20 ratio as a colourless oil. The two isomers were inseperable at this stage of the synthesis.
ÔH (300 MHz, CDCI3): 0.01 [s, 6H, (€ « 3)281], 0.85 [s, 9H, (€ « 3)3081], 1.88 (ddd, J =
14.6, 10.3,1.2 Hz, IH), 1.95 (ddd, J = 14.6,10.3,1.7 Hz, IH), 2.14 (m, 2H), 3.68 (-tt,
J = 10.3, 6.3 Hz, IH, a-08i), 4.70 (brt, J « 5 Hz, IH, a-N), 4.84 (brt, J « 7 Hz, IH, a-O), 5.15 (s, 2H, CH2ph), 6.18 (ddd, J = 9.1, 6.2,1.3 Hz, IH), 6.31 (ddd, J = 9.1, 6.8, 0.8 Hz, IH), 7.32 (m, 5H).
Sc (75 MHz, CDCI3): -4.8 [(€«3)281], 17.9 [(€«3)3081], 25.6 [(€ « 3)3081], 37.2 &
39.8 (2 X C «2), 51.0 (CHN), 66.3 (CHOSi), 67.5 (C«2Ph), 72.3 (CHO), 127.86,127.92 & 128.0 (3 x aryl CH), 128.5 & 129.4 (2 x CH=), 136.0 (aryl C), 156.0 (0=0).
Minor signals corresponding to the 3a-isomer (306) were observed in the NMR spectra of (308).
Vjnax (CH2CI2): 2930m, 2860m, 1690s, 1450brm, 1380m, 1350m, 1270brm, 1230w, 1220w, 1080s, 1005w, 940w, 925w, 905s, 875m, 860w, 835s, 740vbrm cm'^
% (%): 389 (M+, 2), 288 (4), 167 (14), 91 (100).
C2iH3iN04Si [M"*"] requires % 389.2022; observed 389.2022.
lB-Hvdroxv-4B-[(benzvloxvcarbonvl)aminol-6a-[(t-butvldimethvlsilvl)oxv1- cvclohept-2-ene (309)A 9:1 mixture of (306):(308) (5.24 g, 0.013 mol) was reduced with freshly prepared^^ and powdered sodium amalgam (70 g) and sodium phosphate (11.20 g, 0.079 mol) using the procedure described previously for the conversion of (222) into (223). The
residual oily solid obtained was purified by flash chromatography, using 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C), to afford (309) (4.49 g, 85%) as a white solid which had m.p. 93 - 100°C after recrystallisation from toluene and petroleum ether (b.p. 60 - 80°C).
195
§H (300 MHz, CDCI3): 0.11 [s, 6H, (€ « 3)28!], 0.95 [s, 9H, (CH3)3CSi], 1.67 - 1.88 (brm, 3H), 1.93 (brm, IH), 2.47 (brs, IH, exch), 4.24 (brm, IH, a-08i), 4.67 (brm, IH, a-N), 4.79 (brd, J « 9.2 Hz, IH, a-OH), 5.11 (brs, 2H, CH2ph), 5.16 (brm, IH, HN), 5.59 (brd, J « 12Hz, IH), 5.80 (brd, J « 12 Hz, IH), 7.35 (m, 5H).
ÔC (75 MHz, CDCI3): -4.9 [(€«3)281], 18.0 [(€«3)308!], 25.8 [(CH3)3C8i], 41.5 &43.5 (2 X CH2), 45.9 (CHN), 65.5 (CH08Î or CHOH), 66.5 (C«2Ph), 66.7 (CH08i or CHOH), 127.9 (2 x aryl CH), 128.4 (aryl CH), 131.7 (CH=), 136.4 (aryl C), 137.5 (CH=), 155.3 (C=0).
Minor signals coiTesponding to the 6p-isomer (318) were observed. These values are quoted in full later.
Vmax (CH2CI2): 3600W, 3440w, 3040w, 2950m, 2930m, 2880w, 2850w, 1720s, 1505m, 1420brm, 1265brm, 1230m, 1085m, 1040m, 1005w, 940w, 895w, 870w, 835m cm '\
"*/z (%): 373 (M+ - «%€, 1), 335 (14), 334 (36), 290 (20), 272 (10), 208 (11), 183 (17), 91 (100).
Found: C, 64.22; H, 8.39; N, 3.63%.
C2i« 33N048i requires: C, 64.41; H, 8.49; N, 3.58%.
lB-Hvdroxv-2a,3a-epoxv-4B-[(benzvloxvcarbonvl)aminol-6a-[(r-butvldimethvl- silvDoxvlcvcloheptane (310) and
13-hvdroxv-2B,3B-epoxv-4B-[(benzvloxvcarbonvl)amino]-6a-[(r-butvldimethvl- silvDoxvlcvcloheptane (311)A solution of (309) (2.674 g, 6.84 mmol) was epoxidised with 1.2 equivalents of MCPBA (50 - 60% purity) using the procedure described for the conversion of (150) into (255). The resultant crude oil was purified by flash chromatography, eluting with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C), to afford firstly the 2p,3p-epoxide (311) (1.881 g, 68%). An analytical sample was prepared by recrystallising from from toluene and petroleum ether (b.p. 60 - 80°C) to give a white solid (m.p. 114 - 120°C).
ÔH (300 MHz, CDCI3): 0.07 [s, 3H, (€ « 3)28!], 0.09 [s, 3H, (€ « 3)38!], 0.92 [s, 9H,
196
(CHg)3CSi], 1.59 (brd, J « 12 Hz, IH), 1.67 (brd, J= 12 Hz, IH), 1.78 (m, IH), 1.96 (m, IH), 3.18 (brs, IH, exch), 3.23 (d, J = 5.0 Hz, IH, HCO), 3.32 (d, J = 5.0 Hz, IH, HCO), 3.96 (m, IH, a-OSi), 4.37 (brd, J « 8Hz, IH, a-N), 4.47 (brt, J « 9 Hz, IH, a-OH), 5.08 & 5.14 (ABq, J = 12.0 Hz, 2H, CHjPh), 5.64 (brd, J = 9.2 Hz, IH, HN), 7.34 (m, 5H).
Sc (75 MHz, CDCI3): -5.2 [(€«3)281], 17.9 [(€«3)3081], 25.7 [(CH3)gC8i], 38.1 &40.0 (2 X C «2), 45.5 (CHN), 58.5 & 60.1 (2 x CHO), 64.8 (CHOSi or CHOH), 65.5 (CHOSi or CHOH), 66.5 (CH2Ph), 127.9 (2 x aryl CH), 128.3 (aryl CH), 136.3 (aryl C), 155.0 (C=0).
Vmax (CH2CI2): 3600w, 3440W, 3040m, 2950m, 2930m, 2890w, 2860w, 1720s, 1510m, 1420m, 1245m, 1150w, 1095m, 1070m, 1030m, lOOOw, 890m, 835m, 730vbrm cm'^
™/z (%): 407 (M+, 0.5), 350 (31), 306 (19), 91 (100).
Found: C, 61.78; H, 8.25; N, 3.55%.C2iH33N05Si requires: C, 61.88; H, 8.16; N, 3.44%.
Further elution with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) afforded the 2a,3a-epoxide (3 1 0 ) (665 mg, 24%) as a white solid which had m.p. 118 - 123°C after recrystallisation from toluene and petroleum ether (b.p. 60 - 80°C).
Sh (300 MHz, CDCI3): 0.07 [s, 6H, (€ « 3)381], 0.89 [s, 9H, (CH3)3CSi], 1.66 - 2.12 (series of brm, 4H), 2.96 (brs, IH, exch), 3.13 (t, J = 7.5 Hz, IH, HCO), 3.18 (brm, IH, HCO), 3.93 (brm, IH, a-OSi), 4.14 (brm, 2H, a-OH & a-N), 5.13 (brs, 2H, CHzPh), 5.41 (brd, J - 8 Hz, IH, HN), 7.36 (m, 5H).
Sc (75 MHz, CDCI3), The signals quoted in italics were broadened: -5.1 [(€«3)281],17.7 [(€«3)3081], 25.8 [(€ « 3)3081], 39.3 & 40.7 (2 x € « 2), 47.8 (CHN), 56.6 & 58.7 (2 X CHO), 64.6 (CHOSi), 66.6 (CH2Ph), 67.0 (CHOSi), 127.9 (2 x aryl CH), 128.3 (aryl CH), 136.2 (aryl C), 155.5 (0=0).
Vmax (CH2CI2): 3600W, 3430W, 3040m, 2950m, 2930m, 2850m, 1720m, 1510m, 1440m, 1240m, 1145w, 1085w, 1060w, 1040w, 1025w, 1015w, 920w, 890m, 875w,
197
835m, 730brm cm"\
% (%): 407 (M+, 1), 351 (9), 350 (37), 306 (18), 91 (100).
C2iHg3NOgSi [M^] requires ™/z 407.2128; observed 407.2125.
Found: C, 62.23; H, 8.14; N, 3.28%.C2iH33NOgSi requires: C, 61.88; H, 8.16; N, 3.44%.
The NMR spectra of both (3 1 0 ) and (311) were contaminated with minor quantities of the 6P-isomers (319) and (3 2 0 ). As the NMR spectra were broad, a ratio of 3:7 was determined for (3 1 0 ):(3 1 1 ) from the isolated yields.
l3-rp-Toluenesulphonvl)oxv1-2B.3B-epoxv-4B-r(benzvloxvcarbonvl)aminol-6oc- r(t-butvldimethvlsilvl)oxv1cvcloheptane (312 )
A solution of (311) (1.280 g, 3.14 mmol) in THF (35 ml) was tosylated by the sequential addition of n-butyllithium (2.5M in hexane, 1.51 ml, 3.78 mmol) and p-toluenesulphonyl chloride (781 mg, 4.09 mmol) in THF (5 ml) using the procedure described for the tosylation of (244). The tosylate (312) (1.408 g), 85%) was isolated as a foam after flash chromatography, using 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C). NMR analysis indicated that (312) was uncontaminated with minor isomers.
ÔH (300 MHz, CDCI3): -0.01 [s, 3H, (€ « 3)281], 0.04 [s, 3H, (€ « 3)281], 0.90 [s, 9H, (€ « 3)3081], 1.62 (brm, IH), 1.71 - 2.01 (series of brm, 3H), 2.47 (s, 3H), 3.18 (d, J =5.0 Hz, IH, HCO), 3.25 (d, J = 5.0 Hz, IH, HCO), 3.94 (m, IH, a-08i), 4.42 (m, IH, a-N), 5.02 - 5.23 (m, 4H, NH, CH2Ph & a-0802Ar), 7.35 (m, 7H, aryl & part of AA'BB'), 7.82 (2H, part of AA'BB').
6c (75 MHz, CDCI3): -5.5 & -5.2 [2 x (€ « 3)281], 17.9 [(€«3)3081], 21.6 (C«3Ar),25.7 [(€ « 3)3081], 37.5 & 37.2 (2 x € « 2), 45.1 (CHN), 56.9 & 57.4 (2 x CHO), 64.1 (CH081), 66.7 (CH2Ph), 77.5 (CHO8O2), 127.7, 127.95, 128.04, 128.4 & 130.0 (5 x aryl CH), 133.5 (aryl CMe), 136.1 (aryl € € « 2), 144.8 (aryl C8O2), 155.3 (0=0).
Vmax (CH2CI2): 3430W , 3040W, 2950s, 2930s, 2830s, 1725s, 1600w, 1505s, 1440brm, 1360s, 1235s, 1185s, 1175s, 1100s, 1065s, 1040s, 1025w, 1015w, 1005w, 935s, 900s, 870m, 835s, 810m, 725brm cm \
198
™/z (%); 504 [M+ - (CR^hC, 6 ], 350 (6), 306 (4), 229 (18), 91 (100).
la-Chloro-2B,3B-epoxv-4B-r(benzvloxvcarbonvl)aminol-6K-r(?-butvIdimethvlsilvI)- oxvlcvcloheptane (313)A solution of (312) (1.402 g, 2.50 mmol) in DMSO (27 ml) was stirred at 60°C with lithium chloride (690 mg, 16.42 mmol) using the procedure described for the conversion of (245) into (246). The chloride (313) (681 mg, 64%) was isolated as an oil after flash chromatography, eluting with 1:9 diethyl ether:petroleum ether (b.p. 40
- 60°C).
8h (300 MHz, CDCI3): 0.08 [s, 6H, (€ « 3)28]], 0.92 [s, 9H, (€ « 3)3081], 1.66 (m, IH),2.00 (m, IH), 2.17 (m, 2H), 3.33 (m, IH, HCO, p-N), 3.45 (t, J = 4.6 Hz, IH, HCO, P-Cl), 4.02 (m, IH, a-N), 4.19 (m, IH, a-08i), 4.56 (m, IH, a-Cl), 5.06 (brm, IH, HN), 5.15 (s, 2H, CHzPh), 7.38 (m, 5H).
ÔC (75 MHz, CDCI3): -5.0 & -4.9 [2 x (€ « 3)28]], 17.9 [(€«3)3081], 25.7 [(€ « 3)3081], 36.7 & 42.2 (2 x € « 3), 45.0 (CHN), 54.2 (CHCl), 58.4 & 60.5 (2 x CHO), 66.0 (CHOSi), 66.8 (CH2Ph), 128.07, 128.12 & 128.5 (3 x aryl CH), 136.3 (aryl C), 155.3 (0=0).
Vmax (CH2CI2): 3430W, 3040w, 2950m, 2930m, 2880w, 2850m, 1720s, 1505m, 1465w, 1455w, 1435w, 1420w, 1390w, 1360w, 1335w, 1275m, 1240m, 1220m, 1090m, 1065m, 1025w, 1005w, 940w, 905s, 865w, 835m, 725vbrm cm"\
% (%): 425 (M+, 1), 370 (26), 369 (20), 368 (45), 326 (8), 324 (18), 308 (2), 306 (5), 193 (4), 191 (12), 107 (14), 91 (100).
CziHszNO^ClSi [M"""] requires ™/z 425.1789;observed 425.1789.
N-Benzvloxvcarbonvl-3a-[(t-butvldimethvIsilvl')oxvT6B,7B-epoxv-8-azabicvclo- [3.2.11octane (314)Sodium hydride (60% dispersion in mineral oil, 146 mg, 3.65 mmol) was slurried with THF:DME (5:1, 2 ml) under a nitrogen atmosphere. A solution of (313) (310 mg, 0.73 mmol) in THF:DME (5:1, 14 ml) was injected and the mixture was stirred at room temperature for 2 hr. A further portion of sodium hydride (41 mg, 1.03 mmol)
199
was added and stimng was continued at 50°C for 1 hr. The reaction was cooled to -78°C and quenched with the minimum quantity of water. Diethyl ether (30 ml) was added and the organic solution was then washed with water ( 2 x 7 ml) and brine (7 ml). After drying over anhydrous magnesium sulphate and filtration, the solvent was distilled under reduced pressure. The crude oil was purified by flash chromatography, eluting with 1:4 diethyl ether:petroleum ether (b.p. 40 - 60°C)^ to afford (314) (233 mg, 82%) as a white solid which had m.p. 53 - 54°C after recrystallisation from petroleum ether (b.p. 40 - 60°C). The chemical shifts quoted in italics were common to both rotamers.
8h (300 MHz, CDCI3): 0.05 [s, 3H, (€ « 3)281], 0.06 [s, 3H, (€ « 3)281], 0.90 [s, 9H, (€ « 3)3081], 1.64 (brdd, J « 15, 2 Hz, IH), 1.69 (brdd, J= 15, 2 Hz, IH), 2.04 (dt, J =
15,4,4 Hz, IH), 2.10 (dt, J « 15,4,4 Hz, IH), 3.53 (d, J = 3.5 Hz, IH, HCO), 3.56 (d, J = 3.5 Hz, IH, HCO), 4.00 (-t, J = 4.5 Hz, IH), 4.42 (brdd, J = 3.5, 2.0 Hz, IH, HCN), 4.50 (brdd, J = 3.5, 2.0 Hz, IH, HCN), 5.15 & 5.18 (ABq, J = 12.6 Hz, 2H, CHzPh), 7.57 (m, 5H).
8c (75 MHz, CDCI3): -5.0 [(€«3)381], 17.7 [(€«3)3081], 25.7 [(€ « 3)3081], 34.6 &
34.9 (2 X CH2), 53.2 & 53.4 (2 x CHO), 53.7 (2 x CHN), 63.4 (CH081), 66.7 (CH2Ph). 127.7,127.8 & 128.4 (3 x aryl CH), 136.6 (aryl C), 156.6 (0=0).
Vmax (CH2CI2): 3040W, 2950m, 2930m, 2890w, 2860w, 1700s, 1430m, 1360m, 1270brm, llOOw, 1080s, 1040m, 1025w, 1005w, 970w, 940w, 910m, 890m, 840s, 740brmcm'^
™/z (%): 389 (M+, 8), 288 (19), 283 (11), 260 (10), 254 (10), 226 (29), 198 (11), 143 (11), 92 (26), 91 (100).
C2iH3iN048i [M+] requires ™/z 389.2022; observed "'/z 389.2022.
Found: C, 64.53; H, 7.93; N, 3.51%.C2iH3iN04Si requires: C, 64.74; H, 8.02; N, 3.60%.
8copine t-butvldimethvlsilvl ether (315)A solution of (314) (88 mg, 0.23 mmol) in diethyl ether (6 ml) and LAH (45 mg, 1.18 mmol) were gently refluxed under a nitrogen atmosphere for 2 hr. Excess hydride
200
was destroyed by the dropwise addition of water-saturated diethyl ether and the solution was dried over anhydrous sodium sulphate. Filtration and distillation of solvent under reduced pressure gave a residual oil which was purified by flash chromatography, eluting firstly with diethyl ether to remove benzyl alcohol, and secondly with 9:1 ethyl acetate:triethylamine to afford (315) (58 mg, 95%) as a colourless oil.
Sh (250 MHz, CDCI3): 0.00 [s, 6H, (€ « 3)281], 0.86 [s, 9H, (€H3)3€ 8i], 1.47 (dd, J =14.6, 1.1 Hz, 2H), 2.02 (dt, J = 14.6 Hz, J » 4 Hz, 2H), 2.51 (s, 3H), 3.14 (brdd, J « 4 Hz, J= 1.1 Hz, 2H, a-N), 3.62 (s, 2H, HCO), 3.89 (t, J = 4.9 Hz, IH, a-OSi).
Sc (63 MHz, CDCI3): -4.6 [(€«3)281], 18.1 [(€«3)3081], 26.6 [(€ « 3)3081], 35.1 (2 x € « 2), 42.5 (CH3), 57.7 (2 x CHN), 59.0 (2 x CHO), 64.1 (CHOSi).
Vmax (CH2CI2): 3030w, 2930s, 2890s, 2850s, 2795w, 1460m, 1450m, 1390w, 1360w, 1330m, 1240m, 1200m, 1075s, 1040s, 1020s, 1005s, 960w, 935w, 875w, 860s, 840s, 830s, 795m, 720brm cm'^
™/z (%): 269 (M+, 28), 255 (35), 240 (11), 226 (12), 213 (11), 199 (18), 198 (20), 155 (13), 143 (22), 138 (55), 110 (16), 94 (47), 80 (26), 75 (100).
Ci4«27^02Si [M" ] requires "*/z 269.1811; observed "'/z 269.1811.
Scopine (6)TBAF (IM in THF, 0.64 ml, 0.64 mmol) was injected into a solution of (315) (58 mg, 0.22 mmol) in THF (3 ml) under a nitrogen atmosphere. The solution was stirred for 18 hr and the bulk of the solvent was distilled under reduced pressure. The residual oil was dissolved in chloroform (6 ml) and washed with potassium carbonate solution (10 weight %, 2 ml) and brine (2 ml). The organic layer was dried over anhydrous magnesium sulphate, filtered, and the solvent distilled under reduced pressure. The oil was purified by flash chromatography eluting firstly with 1:4 triethylamine:ethyl acetate and then with 1:1:8 methanol:triethylamine:ethyl acetate. The latter fraction afforded scopine (9) (27 mg, 81%) as a white solid which had m.p. 64 - 66°C (literature '^^ m.p. 72°C).
Sh (250 MHz, CDCI3): 1.72 (brdd, J = 15.1 Hz, J = 1.5 Hz, 2H), 2.30 (ddd, J = 15.1,
201
5.3, 4.1 Hz, 2H), 2.35 (brs, IH, exch), 2.71 (s, 3H), 3.38 (brdd, J = 4.1 Hz, J = 1.5 Hz, 2H, a-N), 3.86 (s, 2H, HCO), 4.20 (t, J = 5.3 Hz, IH, a-OH).
ÔC (63 MHz, CDCI3): 34.1 (2 x CH^), 41.9 (CH3), 57.2 (2 x CHN), 58.6 (2 x CHO),63.4 (CHOH).
Vmax (CH2 CI2 ): 3600W, 3450brw, 3020w, 2930m, 1415w, 1390w, 1265brw, 1205w, 1070m, 1040w, 1020w, 975m, 930w, 905w, 865m, 835 cm'^.
™/z (%): 155 (M+, 93), 138 (21), 126 (20), 112 (74), 110 (63), 108 (26), 94 (100), 82 (91), 70 (64), 68 (62).
CgHi3N02 [M'*’] requires 155.0946;observed "'/z 155.0947.
N-Benzyloxvcarbonvl-3a-hvdroxv-6BJB-epoxv-8-azabicvlor3.2.1 loctane (316)A solution of (314) (72 mg, 0.19 nunol) in THF (4 ml) was stirred with TBAF (IM in THF, 0.46 ml, 0.46 mmol) using the reaction and work-up procedure described for the
preparation of scopine (6) from (315). The alcohol (316) (41 mg, 81%) was isolated as a pale yellow oil after flash chromatography, eluting firstly with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) and then with ethyl acetate. Signals quoted in italics are common to both rotamers (in a 1:1 ratio).
Sh (250MHz, CDCI3): 1.61 (brdd, J = 15.1 Hz, J « 2 Hz, IH), 1.65 (brdd, J= 15.1 Hz,
J « 2 Hz, IH), 2.01 (ddd, J = 15.1, J « 4.5, 4.5 Hz, IH), 2.07 (ddd, J = 15.1, J « 4.5,4.5 Hz, IH), 2.20 (brs, IH, exch), 3.47 (d, J = 3.3 Hz, IH, HCO), 3.50 (d, J = 3.3 Hz, IH, HCO), 4.01 (t, J « 4.5 Hz, lH,a-OSi), 4.32 (brdd, J « 4.5, 2Hz, IH, a-N), 4.40 (brdd, J « 4.5,2 Hz, IH, a-N), 5.06 (s, 2H, CHjPh), 7.28 (m, 5H).
Sc (63 MHz, CDCI3): 34.3 & 34.4 (2 x CH2), 53.4 & 53.5 (2 x CHO or CHN), 53.8 &53.9 (2 X CHO or CHN), 63.4 (CHOH), 67.4 (CH2Ph), 128.1, 128.4 & 128.9 (3 x aryl CH), 136.9 (aryl C), 157.3 (C=0).
Vmax (CH2CI2): 3600w, 2930W, 1705s, 1410w, 1385w, 1375w, 1350w, 1295m, 1260w, 1220w, 1100m, 1080m, 1040m, 985w, 910m, 870w, 845w, 835w cm’ .
™/z (%): 275 (M+, 8), 169 (24), 140 (7), 91 (100).
202
C15H17NO4 [M+] requires ™/z 275.1158;obsei-ved ^!z 275.1158.
Norscopine (317)A solution of (316) (37mg, 0.31 mmol) in absolute ethanol (4 ml) was hydrogenolysed at 1 atmosphere using the procedure described for reducing (194). Norscopine (317) (19 mg, 100%) was isolated as a pale yellow waxy solid.
ÔH (250 MHz, CDCI3): 1.53 (dd, J = 14.9, J = 1.3 Hz, 2H), 2.01 (-dt, J = 14.9, 4.9 Hz,
J « 4Hz, 2H), 2.25 (brs, exch, 2H, OH & NH), 3.20 (brdd, J = 4, 1.3 Hz, IH, a-N), 3.53 (s, 2H, HCO), 3.94 (t, J = 4.9 Hz, a-OH).
Sc (63 MHz, CDCI3): 34.9 (2 x CHj), 52.7 (2 x CHN), 54.9 (2 x CHO), 63.5 (CHOH).
Vmax (CH2CI2): 3610w , 3320vbrw, 3000w, 2950s, 1420w, 1395w, 1350w, 1340w, 1325w, 1225brm, 1080s, 1055m, 1045m, 1015w, 985m, 935w, 915w, 865s, 840s cm’ .
"'/z (%): 141 (M+, 36), 122 (33), 112 (36), 84 (24), 83 (35), 82 (44), 80 (61), 72 (28),70 (77), 69 (79), 68 (100).
C7H11NO2 [M+] requires ^!z 141.0790; observed Wz 141.0790.
lB-Hvdroxv-4B-r(benzvloxvcarbonvllamino1-63-l(t-butvldimethvlsilvl)oxv1- cvclohept-2-ene (318)
An 80:20 mixture of (308):(306) (4.05 g, 0.010 mol) was treated with freshly prepared® and powdered sodium amalgam (65 g) and sodium phosphate (11.30 g ,0.080 mol) using the procedure described previously for the reduction of (222) into (223). The solid isolated was recrystallised from toluene and petroleum ether (b.p. 60 - 80°C) to afford (318) (3.57 g, 75%) as a crystalline white solid (m.p. 50 - 51°C). Despite repetitive recrystallisation the product remained contaminated with the 6a-isomer (309).
Sh (300 MHz, CDCI3): 0.12 [s, 3H, ( ^ 3)281], 0.13 [s, 3H, (CH3)2Si], 0.92 [s, 9H, (CH3)3CSi], 1.69 - 1.91 (brm, 2H), 2.03 (brd, J = 13 Hz, IH), 2.15 (brd, J » 13 Hz, IH), 2.70 (brs, IH, exch), 4.10 (brm, IH, a-OSi), 4.28 (brm, 2H, a-N & a-OH), 5.13
203
(brs, 2H, CH^Ph), 5.54 (brd, J = 7 Hz, IH, HN), 5.61 (brd, J « 12 Hz, IH, HC=), 5.80(brd, J = 12 Hz, IH, HC=), 7.38 (m, 5H).
Sc (75 MHz, CDCI3): -4.9 [ ( (« 3)28!], 17.8 [ ( ^ 3)3081], 25.8 [(€ « 3)3081], 41.8 &
44.8 (2 X € « 2), 47.7 (CHN), 66.6 (C«2Ph), 66.7 (CHOSi), 69.3 (CHOH), 127.88,127.92 & 128.3 (3 x aryl CH), 131.7 & 136.4 (2 x CH=), 136.6 (aryl C), 155.3 (C=0).
Vjnax (CH2CI2): 3600w, 3440w, 3040w, 2930s, 2890w, 2850m, 1715s, 1505s, 1415m, 1360W, 1340w, 1305w, 1245m, 1220m, 1085s, 1025s, 940w, 885m, 835s, 730brs cm"\
% (%): 373 (M+ - H2O, 2), 334 (5), 316 (5), 272 (7), 226 (9), 182 (20), 165 (9), 108 (10), 91 (100).
Found: C, 64.34; H, 8.56; N, 3.62%.C2iH33N04Si requires: C, 64.41; H, 8.49; N, 3.58%.
lB-Hvdroxv-2«,3a-epoxv-4B-[(benzvloxvcarbonvllaminol-63-[(t-butvldimethvlsilyl)- oxvl cycloheptane (319) and
lB-Hydroxy-2B.3B-epoxy-4B-r(benzvloxvcarbonyl)aminol-6B-[(t-butvldimethvlsilvl)- oxvlcycloheptane (320)
A solution of (318) (2.129 g, 5.41 mmol) was epoxidised with 1.2 equivalents of MCPBA (50 - 60% purity) using the procedure described for the epoxidation of (150)
into (255). The resultant crade oil was purified by flash chromatography, eluting with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) to afford firstly the 2a,3a-epoxide (319) (980 mg, 44%). NMR analysis indicated that (319) was contaminated with (311). An analytical sample was prepared by recrystallisation from toluene and petroleum ether (b.p. 60 - 80°) and had m.p. 126 - 129°C, although the impurity was still present after recrystallisation.
ÔH (300 MHz, CDCI3): 0.09 [s, 6H, (CH3)28i], 0.89 [s, 9H, (CH3)3CSi], 1.58 - 1.71 (series of brm, 4H), 3.23 (brs, 2H, CHO), 3.88 (vbrs, exch, IH), 4.13 (brm, 2H, a-N & a-OH), 4.40 (m, IH, a-OSi), 5.08 & 5.14 (brABq, J « 13 Hz, 2H, C «2Ph), 5.82 (brd, J « 8 Hz, IH, HN), 7.34 (s, 5H).
Sc (75 MHz, CDCI3), Signals in italics were broadened: -5.1 [(CH3)2Si], 17.7
204
[(CH3)3CSi], 25.6 [(CH3)3CSi], 37.5 & 39.3 (2 x CHg), 49.1 (CHN), 57.2 & 59.0 (2 x CHO), 66.7 (CHaPh), 69.2 (CHOSi or CHOH), 71.2 (CHOSi or CHOH), 127.9 (2 x aryl CH), 128.3 (aryl CH), 136.1 (aryl C), 155.6 (C=0).
Vmax (CH2CI2): 3600W, 3440m, 3040m, 2920s, 2860s, 1715s, 1500s, 1465m, 1415m, 1335m, 1230s, 1150m, 1065s, 1025s, 1005m, 950m, 930m, 885m, 830s, 730vbrs cm'^.
% (%): 407 (M+, <1), 350 (21), 306 (15), 91 (100).
Found: C, 61.84; H, 8.08; N, 3.29%.C2iH33NOgSi requires: C, 61.88; H, 8.16; N, 3.44%.
Further elution with 1:1 diethyl ether:petroleum ether (b.p. 40 - 60°C) afforded the 2p,3p-epoxide (320) (796 mg, 36%) as a white solid which had m.p. 74 - 78°C after recrystallisation from toluene and petroleum ether (b.p. 60 - 80°C). An estimated ratio of 1:1 was made from the isolated yields of the two epoxides.
ÔH (300 MHz, CDCI3): 0.068 [s, 3H, (CH3)2Si], 0.074 [s, 3H, (€ « 3)28]], 0.88 [s, 9H, (CH3)3CSi], 1.71 - 1.86 (m, 2H), 1.92 (m, 2H), 3.22 (d, J = 5.1 Hz, IH, CHO), 3.29 (d, J = 5.1 Hz, IH, CHO), 3.49 (m, IH, a-N), 4.04 (m, IH, a-OH), 4.13 (m, IH, a-OSi), 5.12 (s, 2H, CHzPh), 5.61 (brd, J = 9 Hz, IH, HN), 7.35 (s, 5H).
ÔC (75 MHz, CDCI3): -5.0 [(CH3)2Si], 17.9 [(CH3)3CSi], 25.7 [(€ « 3)3081], 40.3 &42.7 (2 X CH2), 47.3 (CHN), 58.4 & 60.2 (2 x CHO), 66.6 (CHgPh), 66.7 (CHOSi or CHOH), 67.5 (CHOSi or CHOH), 128.0, 128.1 & 128.3 (3 x aryl CH), 136.1 (aryl C), 155.5 (C=0).
Vmax (CH2CI2): 36 0 0 W , 3 4 3 0 w , 2 9 5 0 s , 2 9 3 0 s , 2 8 9 0 w , 2 8 8 0 m , 1 7 2 0 s , 1 5 1 0 s , 1 4 7 0 w ,
1 4 6 0 w , 1 3 6 0 w , 1330W , 1 2 2 0 s , 1 0 9 0 b r w , 1 0 2 0 s , 9 4 0 w , 9 1 0 w , 8 5 5 m , 8 4 0 s , 7 3 0 b r m
cm‘ .
™/z (%): 407 (M+, <1), 350 (8), 306 (11), 91 (100).
CziHggNOgSi [M+] requires ™/z 407.2128; observed ™/z 407.2124.
205
Found: C, 61.71; H, 7.85; N, 3.25%.C2iH33N05Si requires: C, 61.88; H, 8.16; N, 3.44%.
lB-riP-Toluenesulphonvl)oxvl-2B.33-epoxv-4B-r(benzvloxvcarbonvl)amino1-6B- r(r-butvldimethvlsilvl)oxv1cycloheptane (321)A solution of (320) (768 mg, 1.89 mmol) in THF (14 ml) was tosylated by the sequential addition of n-butyllithium (2.5M in hexane, 0.91 ml, 2.30 mmol) and p-toluenesulphonyl chloride (468 mg, 2.45 mmol) in THF (3 ml) using the procedure described for the tosylation of (244). The tosylate (320) (891 mg, 84%) was isolated
as a white foam after flash chromatography, eluting with 3:7 diethyl ether:petroleum ether (b.p. 40 - 60°C).
5h (250 MHz, CDCI3): 0.00 [s, 3H, (€ « 3)281], 0.03 [s, 3H, (€ « 3)281], 0.88 [s, 9H, (€ « 3)3081], 1.64 - 1.98 (m, 3H), 2.12 (m, IH), 2.53 (s, 3H), 3.25 (d, J = 5.0 Hz, IH, HCO), 3.33 (d, J = 5.0 Hz, IH, HCO), 3.43 (m, IH, a-OSi), 4.15 (m, IH, a-N), 4.84 (dd, J = 12.0, 3.3 Hz, IH, a-OS02Ar), 5.18 (s, 2H, C«2Ph), 5.32 (brd, J « 9 Hz, IH),7.42 (m, 7H, aryl & part of AA’BB’), 7.90 (2H, part of AA’BB’).
6c (67 MHz, CDCI3): -4.6 & -4.7 [2 x (€ « 3)381], 18.3 [(€«3)3081], 22.1 (CH3A1),26.0 [(€ « 3)3081], 40.0 & 40.4 (2 x € « 3), 47.4 (CHN), 57.8 & 58.2 (2 x CHO), 67.4 (CHOSi & CHjPh), 77.4 (CHOSO2), 128.2, 128.5, 128.7, 129.0 & 129.2 (5 x aryl CH), 134.0 (aryl CMe), 136.5 (aryl € € « 3), 145.6 (aryl CSO2), 155.9 (0=0).
V m a x (CH2CI2): 3430w, 3030w, 2950s, 2930s, 2890w, 2860s, 1725s, 1600w, 1510s, 1465m, 1370s, 1360s, 1305w, 1220s, 1190s, 1175s, 1095s, 1045m, 1035m, 1005w, 980w, 945s, 920s, 860s, 835s, 810s, 740vbrs cm'^
% (%, FAB): 562 (MH+, 27), 504 (4), 454 (3), 428 (8), 390 (4), 213 (26).
la-Chloro-2B,3B-epoxv-4B-[(benzvloxvcarbonvl)aminol-6B-[(t-butvldimethvlsilyl')- oxy] cycloheptane (322)
A solution of (321) (891 mg, 1.59 mmol) in DMSO (12 ml) was stirred at 75°C with lithium chloride (414 mg, 9.86 mmol) using the procedure described for the conversion of (245) into (246). The chloride (322) (534 mg, 79%) was isolated as an foam after flash chromatography, eluting with 1:4 diethyl ether:petroleum ether (b.p. 40 - 60°C).
206
ÔH (250 MHz, CDCI3); 0.00 [s, 6H, ( ^ 3)281], 0.80 [s, 9H, ( ^ 3)3081], 1.67 - 1.82 (m, 2H), 2.02 (m, 2H), 3.18 (brdd, J = 8, 4.5 Hz, HCO, p-Cl), 3.22 (d, J = 4.7 Hz, IH, HCO, p-N), 3.92 (m, IH, a-08i), 4.25 (m, IH, a-N), 4.52 (brdd, J « 8, 4.5 Hz, IH, a-Cl), 5.02 (s, 2H, CHjPh), 5.45 (brd, J « 9 Hz, IH, HN), 7.27 (m, 5H).
8c (63 MHz, CDCI3): -4.4 [ ( ^ 3)281], 18.4 [(CH3)3C8i], 26.1 [(CH3)3CSi], 39.7 &
41.8 (2 X CH2), 47.4 (CHN), 56.0 & 57.8 (2 x CHO), 60.5 (CHCI), 66.5 (CH2PI1), 67.3 (CH08i), 128.5,128.6 & 128.9 (3 x aryl CH), 136.8 (aryl C), 156.0 (C=0).
Vmax (CH2CI2): 3430w, 2950m, 2930m, 2890w, 2860w, 1720s, 1505s, 1465w, 1455w, 1390w, 1375w, 1360w, 1330w, 1265brw, 1230m, 1220m, 1175w, 1115w, 1075m, 1020w, 1005w, 940w, 910m, 855m, 835m, 740brw cm"\
"=/z (%): 425 (M+, <1), 370 (4), 368 (11), 326 (2), 324 (6), 149 (5), 91 (100).
C2iH32N048iCl [M+] requires ™/z 425.1789; observed Wz 425.1789.
N-Benzyloxycarbonyl-3 3- r(r-butvldimethvlsilvl)oxvl-6P,7 B-epoxy-8-azabicylo- r3.2.11octane (323)A solution of (322) (154 mg, 0.36 mmol) in THF:DME (5:1, 10 ml) was cyclised with sodium hydride (60% dispersion in mineral oil, 105 mg, 2.63 mmol) using the procedure described for the conversion of (313) into (314). After flash chromatography, eluting with 1:4 diethyl ether:petroleum ether (b.p. 40 - 60°C), (323) (96 mg, 68%) was isolated as a colourless oil. The signals quoted in italics are
common to both rotamers (in a 1:1 ratio).
ÔH (250 MHz, CDCI3): 0.00 [s, 6H, (CH3)28i], 0.83 [s, 9H, (CH3)3C8i], 1.65 (dt, J = 10.2, 10.2, J = 3 Hz, IH), 1.72 (dt, J = 10.2, 10.2, J « 3 Hz, IH), 1.90 (ddd, J = 10.2, 6.7, J « 3 Hz, IH), 1.94 (ddd, J = 10.2, 6.7, J « 3 Hz, IH), 3.34 (d, J = 3.1 Hz, IH, HCO), 3.38 (d, J = 3.1 Hz, IH, HCO), 4.08 (tt, J = 10.2, 6.7, IH, a-OSi), 4.37 (t, J « 3 Hz, IH, a-N), 4.46 (t, J = 3 Hz, IH, a-N), 5.10 (s, 2H, CHgPh), 7.32 (m, 5H).
8c (63 MHz, CDCI3): -4.3 [(CH3)28i], 18.4 [(CH3)3C8i], 26.1 [ (€ « 3)3081], 35.8 &36.1 (2 X CH2), 52.1 & 52.4 (2 x CHO), 53.8 & 54.2 (2 x CHN), 64.5 (CH081), 67.4 (CH2Ph). 128.2,128.4 & 128.9 (3 x aryl CH), 137.0 (aryl C), 156.7 {0=0).
207
Vmax (CH2CI2): 2950m, 2930m, 2895w, 2860m, 1705s, 1505w, 1410brm, 1365w, 1330W, 1295m, 1230m, 1100s, 1080m, 1030w, 1005w, 975w, 940w, 915w, 880m, 860m, 845m, 835m cm"^
™/z (%): 389 (M+, < 1), 332 (37), 288 (5), 256 (8), 182 (68), 91 (100).
C2iH3iN04Si [M+] requires % 389.2022; observed ™/z 389.2021.
Pseudoscopine f-butyldimethylsilyl ether (324)
This compound was prepared by reducing (323) (93 mg, 0.24 mmol) in diethyl ether (5 ml) with LAH (40 mg, 1.05 mmol) using the procedure described for the reduction of (314). The amine was purified by flash chromatography, eluting firstly with diethyl ether and then with ethyl acetate:triethylamine (9:1), to afford (324) (61 mg, 95%) as a colourless oil.
ÔH (250 MHz, CDCI3): 0.00 [s, 6H, ( ( « 3)281], 0.84 [s, 9H, (CH3)3C8i], 1.70 (m, 4H), 2.49 (s, 3H), 3.22 (t, J = 2.9 Hz, 2H, a-N), 3.44 (s, 2H, HCO), 4.00 (tt, J = 9.1, 7.8 Hz, IH, a-08i).
8c (63 MHz, CDCI3): -4.2 [(CH3)28i], 18.4 [(CH3)3C8i], 26.2 [(CH3)3C8i], 33.9 (2 x
CH2), 39.0 (CH3), 55.6 (2 X CHN), 58.7 (2 x CHO), 64.4 (CH08i).
Vmax (CH2CI2): 3030W, 2950s, 2930s, 2890s, 2870s, 1470w, 1475w, 1390w, 1360w, 1335w, 1245w, 1205w, 1165w, 1095s, 1080s, 1030w, 1005w, 980w, 965w, 940w, 875s, 865s, 845s, 835s, 740brm cm \
% (%): 269 (M+, 9), 226 (12), 224 (12), 212 (29), 138 (28), 114 (13), 111 (18), 108 (25), 101 (12), 97 (23), 94 (28), 91 (100).
Ci4H27N028i [M+] requires ™/z 269.1811; observed *"/z 269.1811.
Pseudoscopine (297)
This compound was prepared by treating (324) (61mg, 0.23 mmol) in THF (4 ml) with TBAF (IM in THF, 0.66 ml, 0.66 mmol) using the procedure described to deprotect (315). Pseudoscopine was purified by flash chromatography, eluting firstly
208
with 1:4 triethylamine:ethyl acetate and secondly with 1:3:1 methanohethyl acetate: triethylamine to afford (297) (29 mg, 83%) as a white solid. M.p. 120 - 122°C (lit.^ 121 - 122«C).
5h (250 MHz, CDCI3): 1.83 (ddd, J = 13.5, 9.7 Hz, J = 3.5 Hz, 2H), 2.00 (ddd, J =13.5, 6.7, 2.4 Hz, 2H), 2.66 (s, 3H), 2.90 (brs, exch, IH), 3.41 (dd, J » 3.5 Hz, J = 2.4 Hz, 2H, a-N), 3.63 (s, 2H, HCO), 4.16 (tt, J = 9.7, 6.7 Hz, IH, a-OH).
6c (63 MHz, CDCI3): 33.9 (2 x CHg), 40.1 (CH3), 55.8 (2 x CHN), 58.7 (2 x CHO),63.8 (CHOH).
V m a x (CHgClg): 3620W , 3400brw, 3030w, 2940s, 1520w, 1470w, 1440, 1390w, 1370W , 1335w, 1270brw, 1220w, 1160w, 1140w, 1075m, 1055s, 1030w, 980w, 970w, 960w, 910w, 870s, 845s, 815w cm'^.
™/z (%): 155 (M+, 100), 138 (30), 126 (23), 112 (56), 110 (75), 108 (12), 97 (20), 94 (32), 86 (22), 84 (25), 82 (42), 70 (26), 68 (24), 57 (79).
CgHi3N02 [M+] requires % 155.0946; observed *®/z 155.0946.
N-Benzyloxvcarbonvl-3b-hvdroxv-6B.7B-epoxv-8-azabicvlor3.2.11octane (325)A solution of (323) (81 mg, 0.21 mmol) in THF (5 ml) was stirred with TBAF (IM in THF, 0.52 ml, 0.52 mmol) using the reaction and work-up procedure described for the
preparation of scopine (6) from (315). The alcohol (325) (40 mg, 70%) was isolated as a yellow oil after flash chromatography, eluting with ethyl acetate. Signals quoted in italics aie common to both rotamers (in a 1:1 ratio).
ÔH (250 MHz, CDCI3): 1.55 (ddd, J = 13.6, 10.6 Hz, J « 3 Hz, IH), 1.60 (ddd, J =13.6, 10.6 Hz, J « 3 Hz, IH), 1.99 (m, 2H), 2.28 (brs, IH, exch), 3.30 (d, J = 3.1 Hz, IH, HCO), 3.33 (d, J = 3.1 Hz, IH, HCO), 4.04 (tt, J = 10.6, 6.6 Hz, IH, a-OH), 4.35 (brt, J « 3 Hz, IH, a-N), 4.42 (brt, J » 3 Hz, IH, a-N), 5.04 (s, 2H, CHjPh), 7.27 (m, 5H).
6c (63 MHz, CDCI3): 35.4 & 35.6 (2 x CHg), 52.0 & 52.3 (2 x CHO or CHN), 53.8 &54.1 (2 X CHO or CHN), 63.9 (CHOH), 67.5 (CKJ?h), 128.2, 128.5 & 128.9 (3 x aryl CH), 136.8 (aryl C), 156.8 (C=0).
209
Vmax (CH2CI2): 3600W, 3490brw, 3050w, 3950w, 3930w, 2860w, 1710s, 1500w, 1410brm, 1365m, 1280brm, 1220w, 1100m, 1080m, 1060s, 1025w, 995w, 970w, 910w, 870w, 860w, 855m cm"k
% (%): 275 (M+, 24), 141 (3), 124 (3), 108 (3), 168 (12), 91 (100).
C15H17NO4 [M+] requires "'/z 275.1158; observed '"/z 275.1157.
Norpseudoscopine (326)A solution of (325) (31 mg, 0.11 mmol) in absolute ethanol (5 ml) was hydrogenolysed at 1 atmosphere, using the procedure described for reducing (194) into (197). Norpseudoscopine (326) (13 mg, 82%) was isolated as a white solid which had no specific m.p.
ÔH (250 MHz, CDCI3): 1.60 (ddd, J = 13.4, 9.9 Hz, J = 3.5 Hz, 2H), 1.92 (ddd, J = 13.4, 6.5 Hz, J » 3.5 Hz, 2H), 2.42 (brs, exch, 2H, OH & NH), 3.28 (brt, J « 3.5 Hz, 2H, a-N), 3.33 (s, 2H, HCO), 3.88 (tt, J = 9.9, 6.5 Hz, IH, a-OH).
ÔC (63 MHz, CDCI3): 36.1 (2 x CH2), 53.3 (2 x CHN or CHO), 53.5 (2 x CHN or CHO), 64.1 (CHOH).
Vmax (CH2CI2): 3600w, 3030w, 2950m, 2850w, 1270brw, 1075m, 1055m, 970w, 910m, 865w, 840w, 835w cm'k
'"/z (%): 141 (M+, 18), 124 (37), 122 (35), 112 (73), 98 (51), 97 (70), 96 (79), 94 (42),82 (24), 80 (31), 70 (85), 68 (100).
C7H11NO2 [M+] requires % 141.0790; observed "’/z 141.0790.
N-(Benzvloxycarbonyl)-3a,4a-epoxv-6-aza-7-oxabicvclor3.2.21non-8-ene (329)A small portion of the crade epoxide (301), used previously to prepare (302), was purified by vacuum distillation (18 mbar, 40°C). Tétraméthylammonium periodate (283 mg, 0.89 mmol) was added to a solution of (301) (80 mg, 0.74 mmol) in dichloromethane (7 ml) and stirred at 0°C. A solution of benzyl-N-hydroxycarbamate (148 mg, 0.89 mmol) in dichloromethane (5 ml) was dripped in over 10 min. The
210
mixture was warmed to ambient temperature and stirred for a further 3 hr. The solution was filtered, washed with saturated sodium thiosulphate solution ( 2 x 5 ml), saturated sodium bicmbonate solution (5 ml) and brine (5 ml). The organic layer was dried over anhydrous magnesium sulphate, filtered and the solvent distilled under reduced pressure to afford a crude oil. The oil was purified by flash chromatography,
eluting with 6:4 diethyl ether:petroleum ether (b.p. 40 - 60 °C) to afford a mixture of (3 2 9 ):(3 3 0 ) in an 80:20 ratio (estimated from NMR integrations) as a white solid (130 mg, 64%). The two isomers were inseparable by chromatography but recrystallisation from diethyl ether and petroleum ether (b.p. 60 - 80 °C) furnished (3 2 9 ) in pure form (m.p 86 - 87°C). The stereo- and regiochemistry of (329) was elucidated from a crystal structure (appendix 1).
ôjj (300 MHz, CDCI3), Major cycloadduct (329): 2.32 (m, 2H), 2.14 (m, IH, HCO), 3.37 (dd, J = 6.2, 4.4 Hz, IH, HCO), 4.58 (m, IH, bridgehead a-N), 5.18 & 5.23 (ABq, J = 12.3 Hz, 2H, CH2?h), 5.41 (dt, J = 6.2, 6.2, 1.7 Hz, IH, bridgehead a-0), 5.40 (ddd, J = 9.2, 7.1, 1.1 Hz, IH, HC=), 6.22 (ddd, J = 9.2, 6.2, 1.6 Hz, IH, HC=), 7.36 (m, 5H).
ÔC (75 MHz, CDCI3): 31.4 (CH2), 50.0 & 53.3 (2 x epoxide CHO), 54.6 (CHN), 68.1 (CH2?h), 72.7 (bridgehead CHO), 128.0 (aryl CH), 128.1 (HC=), 128.2 & 128.5 (2 x aryl CH), 129.6 (HC=), 135.6 (aryl CH), 157.3 (C=0).
Umax (CH2CI2): 3060W , 30 4 0 W , 3 0 00W , 2 9 5 0 w , 2 9 2 0 w , 1 7 0 5 s , 1 5 0 0 w , 1 4 4 5 m ,
1 3 9 5 m , 1 3 8 0 m , 1 3 5 0 m , 1 2 7 5 b r s , 1 1 5 5 w , 1 0 9 5 m , 1 0 7 0 m , 1 0 5 0 m , 1 0 4 0 m , 9 8 0 w ,
9 6 0 w , 9 4 5 w , 9 0 5 w , 8 7 0 m , 8 6 0 m cm '^.
% (%): 273 (M+, 2), 229 (6), 108 (7), 107 (7), 92 (18), 91 (100), 79 (13), 97 (10).
Found: C, 65.66; H, 5.36; N, 5.19%.C15H15NO4 requires: C, 65.90; H, 5.53; N, 5.13%.
Additional signals in the NMR spectra (before recrystallisation) corresponded to the minor cycloadduct (330):
ôjj (300 MHz, CDCI3): 4.68 (m, IH, bridgehead a-N), 5.14 (m, IH, bridgehead a-0),6.06 (dd, J = 9.1,7.4 Hz, IH, HC=), 6.33 (m, IH, HC=).
211
ÔC (75 MHz, CDCI3); 29.7 (CH^), 51.1, 51.6 & 54.0 (3 x CH), 67.7 (CH^Ph), 74.7 (bridgehead CHO).
212
References
1. (a) G. Fodor and R. Dharanipragada, Nat. Prod. Rep., 1988,5, 67.(b) G. Fodor and R. Dharanipragada, ibid., 1990,7, 539.(c) G. Fodor and R. Dharanipragada, ibid., 1991,8,603.
(d) G. Fodor and R. Dharanipragada, ibid., 1992,10, 199.(e) G. Fodor and R. Dharanipragada, ibid., 199410,443.
2. (a) H.C.S Wood and R. Wriggleswoith, in Rodds Chemistry of CarbonCompounds, ed. S. Coffey, Elsevier Scientific Publishing Co., 1977, Vol. 4 part B, p. 201.
(b) H.L. Holmes, in The Alkaloids, ed. R.H.F Manske, Academic Press, New York, 1950, Vol. 1, p.271.
(c) G. Fodor, ibid, 1960, Vol. 6, p. 145.
(d) G. Fodor, ibid, 1967, Vol. 9, p. 269.(e) G. Fodor, ibid, 1971, Vol. 13, p. 351.(f) R. L. Clarke, ibid, 1977, Vol. 16, p. 83.(g) M. Lounasmaa, ibid, 1988, Vol. 33, p. 1.(h) M. Lounasmaa and T. Tamminen, ibid, 1994, Vol. 44, p. 1.
3. L. Geiger and C. Hesse, Ann., 1833,5,43; 1833, 6,44; 1833, 7, 269.4. F. Wohler, Ann., 1862,121, 372.5. (a) J. Berberich, Deut. Med. Wochschr, 1931,57,1579.
(b) D.E. Adelson, L.G. MacDowell and C.B. Pollard, J. Am. Chem. Soc., 1935,57,1988.
6. R. Willstatter and E. Berner, Chem. Ber., 1923,56, 1079.7. Bao Gong Teng Cooperative Research Group, Zhongcaoyao, 1982,13,20.8. T. Yao, Z. Chen, D. Yi and G. Xu, Yaoxue Xuebao, 1981,16, 582; Chem.
Abstr. 1982, 96,48972c.
9. P. Wang, T. Yao and Z. Chen, Huaxue Xuebao, 1989,47, 1002; Chem. Abstr. 1990,113,78746u.
10. (a) Z. Xiang, J.E. Zhou, Z.N. Chen, L.P. Wang, H.N Wang, T.R. Yao,J.X. Xie, G.Y. Xu and D.N. Yi, Yaoxue Xuebao, 1989,24,105; Chem. Abstr. 1990,112,198846c.
(b) M.E. Jung, Z. Longmei, P. Tangsheng, Z. Huiyan, L. Yan and S. Jingyu, J. Org. Chem., 1992,57,3528.
11. T.F. Spande, H.M. Ganaffo, M.W. Edwards, H.J.C Yeh, L. Pannell and J.W. Daly, J. Am. Chem. Soc., 1992,114, 3475.
12. C.A. Broka, Tetrahedron Lett., 1993, 34, 3251.13. R.R. Fraser and R.B. Swingle, Can. J. Chem., 1970,48,2065.
213
14. S.R. Fletcher, R. Baker, M.S. Chambers, S.C. Hobbs and P.J. Mitchell,J. Chem. Soc., Chem. Commun., 1993, 1216.
15. E.J. Corey, T-P. Loh, S. Achyutharao, D C. Daley and S. Sarshar, J. Org. Chem., 1993,58,5600.
16. (a) D.F. Huang and T.Y. Chen, Tetrahedron Lett., 1993,34,4477.(b) S.C. Clayton and A.C. Regan, Tetrahedron Lett., 1993, 34, 7493.(c) C. Szantay, Z. Kardos-Balogh, I. Moldvai, C. Szantay Jr.,
E. Temesvari-Major and G. Blasko, Tetrahedron Lett., 1994, 35,3171.
(d) G. Pandey, T.D. Bagul and G. Lakshmaiah, Tetrahedron Lett., 1994, 35,7439.
(e) E. Albertini, A. Baico, S. Benetti, C. De Risi, G.P. Pollini,R. Romagnoli and V. Zanirato, Tetrahedron Lett., 1994, 35, 9297.
(f) S.Y. Ko, J. Leipiniere, I.D. Linney, and R. Wrigglesworth, /. Chem. Soc., Chem. Commun., 1994, 1775.
17. R. Robinson., J. Chem. Soc., 1917, 111, 762.18. C. Schopf and G. Lehmann, Ann., 1935,518, 1.19. A. Stoll, B. Becker and E. Jucker, Helv. Chim. Acta., 1952, 3 5 ,1263.20 R.C. Menzies and R. Robinson, J. Chem. Soc., 1924,125, 2163.21. C. Schopf and A. Schmetterling, Angew. Chem., 1952,64, 591.22. A.T. Bottini and J. Gal, J. Org. Chem., 1971,36,1718.23. Y. Kashman and S. Cherkez, Synthesis, 1974, 885.24. R. Willstatter and A. Pfannenstiel, Ann., 1922,422,1.25. G. Cignarella, G.G. Gallo and E. Testa, J. Am. Chem. Soc., 1961, 43, 4999.26. J.B. Bapat, D.St.C. Black, R.F.C. Brown and C. Ichlov, Aust. J. Chem.,
1972,25,2445.27. (a) J.J. Tufariello and E.J. Trybulski, J. Chem. Soc., Chem. Commun.,
1973,720.(b) J.J. Tufariello, G.B. Mullen, J.J. Teegler and E.J. Trybulski, J. Am.
Chem. Soc., 1979,101, 2435.28. (a) N.J. Tuito and S.S. Edelson, J. Am. Chem. Soc., 1968,90,4499.
(b) N.J. Tuito, S.S. Edelson, J.R. Williams, T.R. Darling andW.B. Hammond, J. Am. Chem. Soc., 1969, 91,2283.
29. G. Fierz, R. Chidgey and H.M.R. Hoffmann, Angew. Chem., 1974, 86,44430. Y. Hayakawa, Y. Baba, S. Makino and R. Noyori, J. Am. Chem. Soc., 1974,
100, 1786.
214
31. J. Mann and L.C. de A. Barbosa, J. Chem. Soc., Perkin Trans. 1, 1992,787.
32. (a) H.M.L. Davies, W.B. Young and H.D. Smith, Terra/îeJronLert.,1989,30,4653.
(b) H.M.L. Davies, E. Saikali and W.B. Young, J. Org. Chem., 1991, 56, 5696.
33. G. Pandey, G. Lakshmaiah and A. Ghatak, Tetrahedron Lett., 1994, 34, 7301.
34. (a) H. lada, Y. Watanabe and C. Kibayashi, Tetrahedron Lett., 1984,25,5091.
(b) H. lada, Y. Watanabe and C. Kibayashi, J. Org. Chem., 1985, 50, 1818.
35. (a) A. Bathgate and J.R. Malpass, Tetrahedron Lett., 1987,28, 5937.(b) A. Naylor (née Bathgate), N. Howarth and J.R.Malpass, Tetrahedron,
1993,49,451.(c) A. Bathgate, Syntheses and Reactions of Azabicycles, Ph.D. Thesis,
University of Leicester, England, 1988.36. J.E. Backvall, Z.D. Renko and S.E. Bystrom, Tetrahedron Lett., 1987,28,
4199.37. H.E. Schink, H. Petterson and J.E. Backvall, J. Org. Chem., 1991, 56,2769.38. W.W. Carmichael, D.F. Biggs and P.R. Gorham, Science, 1975,187, 542.39. O.M. Skulberg, W.W. Carmicheal, R.A. Anderson, S.Matsunaga,
R.E. Moore and R. Skulberg, Environ. Toxicol. Chem., 1992,11, 321.40. C.S. Huber, Actacrystallogr. Sect. B., 1972, 78, 2577.
41. J.P. Devlin, G.E. Edwards, P.R. Gorman, N.R. Hunter, R.K. Pike andB. Stavric, Can. J. Chem., 1977,55,1367.
42. (a) W.W. Carmicheal, D.F. Biggs and M.A. Peterson, Toxicon, 1979,17,229.
(b) C.E. Spivak, B. Witcop and E.X. Albuquerque, Mol. Pharmacol., 1980,18, 384.
(c) C.E. Spivak, J. Waters, B. Witkop and E.X. Albuquerque, Mol. Pharmacol., 1983, 23, 337.
43. H.F. Campbell, G.E. Edwards and R. Kolt, Can. J. Chem., 1977, 55, 1372.44. H.A. Bates and H. Rapaport, J. Am. Chem. Soc., 1979,101,1259.45. J.S. Peterson, G. Fels and H. Rapaport, J. Am. Chem. Soc., 1984,106, 4539.46. J.J Tufariello, H. Meckler, K. Pushpananda and A. Sanarante, J. Am.
Chem. Soc., 1984,106,7979.
215
47. R.L. Danheiser, J.M. Morin Jr. and E.J. Salaski, J. Am. Chem. Soc., 1985, 107, 8066.
48. (a) K.H. Melching, H. Hiemstra, W.J. Klaver and W.N. Speckamp,Tetrahedron Lett., 1986,27,4799.
(b) J.R. Wiseman and S.Y. Lee, J. Org. Chem., 1986,51, 2485.
(c) P.M. Esch, H. Hiemstra, W.J. Klaver and W.N. Speckamp, Heterocycles, 1987,26,75.
(d) T. Shono, Y. Matsumura, K. Uchida and K. Tagami, Chemistry Lett., 1987,919.
(e) P. Vernon and T. Gallagher, J. Chem. Soc., Chem. Commun., 1987, 245.
49. P. Somfai and J. Âhman, Tetrahedron Lett., 1992, 33, 3791.50. P. Somfai, H.M.He and D. Tanner, Tetrahedron Lett., 1991,32, 283.51. A.C. Cope, H.R. Nace and L.L. Esters Jr., J. Am. Chem. Soc., 1950, 72,
1123.
52. (a) A.G. Anastassiou, J. Am. Chem. Soc., 1965, 87, 5512.(b) A.G. Anastassiou, J. Am. Chem. Soc., 1968, 90, 1527.
53. A.G. Anastassiou and R.P. Cellura, J. Org. Chem., 1972,37, 3126.54. J.W. Bastable, J.D. Hobson and W.D Riddell, J. Chem. Soc., Perkin Trans.
7, 1972,2205.55. (a) G. Haufe and E. Kleinpeter, Tetrahedron Lett., 1982,23, 3555.
(b) G. Haufe, U. Rolle, E. Kleinpeter, J. Kivikoski and K. Rissanen,J. Org. Chem., 1993, 58, 7084.
56. M.A. Loreto, L. Pellacani and P.A. Tardella, J. Heterocycl. Chem.,1979,16, 1233.
57. J. Barluenga, J. Pérez-Prieto and G. Asensio, J. Chem. Soc., Chem. Commun., 1982, 1181.
58. (a) C.R. Smith and J.R. Malpass, Tetrahedron Lett., 1992,33, 273(b) C.R. Smith, D.E. Justice and J.R. Malpass, Tetrahedron, 1993,49,
11037.(c) C.R. Smith, Homotropanes - Synthetic Approaches to the
9-Azabicyclo[4.2.1 Jnonane/ene Ring System, Ph.D. Thesis, University of Leicester, England, 1992.
59. (a) M.E. Christy, P.S. Anderson, S.E. Britcher, C D. Colton, B.E. Evans,D C. Remy and E. Engelhardt, J. Org. Chem., 1979, 44, 3117.
216
(b) T.R Lamanec, D.R. Bender, A.M. DeMarco, S. Karady, R.A.
Reamer, and L.W. Weinstock, J. Org. Chem., 1988, 53, 1768.60. P.D. Leeson, K. James and R. Baker, J. Chem. Soc., Chem. Commun.,
1989,433.61. (a) K.E.Harding and S.R. Burks, J. Org. Chem., 1981,46, 3920.
(b) K.E. Harding and M.W. Jones, Heterocycles, 1989,28, 663.62. (a) K.E.Hai'ding and S.R. Burks, J. Org. Chem., 1984,49,40.
(b) S. Danishefsky, E. Taniyama and R.R. Webb II, Tetrahedron Lett., 1983,24,11.
63. J.M Takacs, M.A. Helle and F. Takusagawa, Tetrahedron Lett., 1983, 24, 11.
64. K.E Haiding and T.H. Marman, /. Org. Chem., 1984, 49, 2838.65. (a) C.E. Miller, J. Chem. Educ., 1965,42,1965,
(b) W. Adam, O. Cueto, O. De Lucci, K. Peters, E-M. Peters and H.G. Von Schnering, J. Am. Chem. Soc., 1981,103, 5822.
66. J. W. Hamersma and E.I Snyder, J. Org. Chem., 1965, 30, 3985.67. (a) S.F. Martin, M. Hartmann and J.A. Josey, Tetrahedron Lett., 1992,
33,3583.(b) Fieser and Fieser, Reagents for Organic Synthesis, J. Wiley and Sons,
1967, Vol. 1, p. 1030.68. (a) K. Bowden, I.M. Heibron, E.R.H Jones and B.C.L. Weedon, J. Chem.
Soc., 1946, 39.(b) A. Bowers, T.G Halsall, E.R.H. Jones and A.J. Lemin, J. Chem. Soc.,
1953, 2548.69. R. Greenwald, M. Chaykovsky and E.J. Corey, J. Org. Chem., 1963, 28,
1128.70. G. Wittig and U. Schollkopf, Org. Syn., 1960,40, 66.71. (a) G.A. Gray and W.R. Jackson, J. Am. Chem. Soc., 1969, 91, 6205.
(b) B. Giese and J. Meister, Chem. Ber., 1977,110, 2588.
(c) J. Barluenga and M. Yus, Chem. Rev., 1988, 88,487.72. G.M. Whitesides and J. San Filippo, Jr., J. Am. Chem. Soc., 1970, 92, 6611.73. R.P. Quirk, J. Org. Chem., 1972, 37, 3554.74. (a) F.H. Senbold, J. Am. Chem. Soc., 1953, 75, 25332.
(b) E.J. Hamilton Jr., and H. Fischer, Helv. Chim. Acta., 1973,56,795.75. C.L. Hill and G.M. Whitesides, J. Am. Chem. Soc., 1974, 96, 870.76. A. Bottini, Selective Organic Transformations, ed. B.S. Thyagarajan,
217
Wiley Interscience, New York, 1970, p. 89.77. (a) R.H. Shapiro and M.J. Heath, /. Am. Chem. Soc., 1967, 89, 5734.
(b) R.H. Shapiro, Org. React. (N. Y.), 1975, 23, 405.(c) R.H. Shapiro, M.F. Lipton, K.J. Kolonko, R.L. Buswell and
L.A. Capuan, Tetrahedron Lett., 1975, 1811.78. J. Casanova, B. Waegell, G. Kouokoua and V. Toure, J. Org. Chem.,
1979,44, 3976.79. J. Barluenga, J. Perez-Prieto, G. Asensio, S. Garcia-Granda and
M.A. Salvado, Tetrahedron, 1992, 48, 3813.80. J. Attenburrow, A.F.B. Cameron, J.H. Chapman, R.M. Evans, B.A. Hems,
A.B.A. Jansen and T. Walker, J. Chem. Soc., 1952,1094.81. H. Firouzabadi and E. Ghaderi, Tetrahedron Lett., 1978, 839.82. (a) H.C. Brown and P.J. Geoghegan Jr., J. Org. Chem., 1970, 35, 1844.
(b) H.C. Brown, P.J. Geoghegan Jr., G.J Lynch and J.T. Kurek, /. Org.Chem., 1972,37,1941.
83. (a) T.G Traylor and A.W. Baker, Tetrahedron Lett., 1959, 14.(b) T.G Traylor and A.W. Baker, J. Am. Chem. Soc., 1963,85,2746.
84. (a) G. Fodor, J. Toth, I. Koczor, P. Dobo and I. Vincze, Chem.Ind. (London), 1956, 764.
(b) P. Dobo, G. Fodor, G, Janzso, I. Koczor, J. Toth and I. Vincze,J. Chem. Soc., 1959, 3461.
85. P. Kocovsky and I. Stary, J. Org. Chem., 1990,55, 3236.86. H.C. Brown, J.H. Kawakami and S. Ikegami, /. Am. Chem. Soc., 1970, 92,
6914.87. N. Howarth, Synthetic Approaches to Derivatives o f Tropane, Ph.D Thesis,
University of Leicester, England, 1991.
88. (a) A.B. Ray, M. Sahai and P.D. Sethi, Chem. Ind. (London), 1976,454. (b) A.B. Ray and M. Sahai, J. Org. Chem., 1980,45, 3265.
89. A.R. Pinder, /. Org. Chem., 1982,47, 3607.90. (a) A.B. Ray, Y. Oshima, H.Hikino and C. Kabuto, Heterocycles, 1982,
19,1233.(b) A.T. McPhail and A.R. Pinder, Tetrahedron, 1984,40, 1661.
91. D. Tepfler, A. Goldmann, N. Pamboukdian, M. Maille, A. Lepingle, D.Chevalier, J. Denarié and C. Rosenbury, J. ofBact., 1988,170, 1153.
92. (a) P-H. Ducrot and J.Y Lallemand, Tetrahedron Lett., 1990, 31, 3879. (b) P-H. Ducrot and J.Y Lallemand, ibid, 1990,31, 3883.
218
93. (a) C.R. Smith and J.R. Malpass, Tetrahedron Lett., 1992, 33, 277.(b) C.R. Smith, D.E. Justice and J.R. Malpass, Tetrahedron, 1994, 50,
11039.94. R.S. Lott, V.S. Chauhan and C.H. Stammer, J. Chem. Soc., Chem.
Commun., 1979,495.95. E.J. Corey and G. Schmidt, Tetrahedron Lett., 1979, 399.96. (a) E.J. Corey and J.W. Suggs, Tetrahedron Lett., 2647,1975.
(b) E.J. Corey and D.L. Boger, ibid., 2461,1978.97. W.P. Griffith, S.V. Ley, G.P. Whitcombe and A.D. White, J. Chem. Soc.,
Chem. Commun., 1987,1625.98. D.E. Justice and J.R. Malpass, J. Chem. Soc., Perkin Trans., 1994, 2559.99. D. Belkacemi, J.W. Davies, J.R. Malpass, A. Naylor (née Bathgate) and
C.R. Smith, Tetrahedron, 1992, 48, 10161.100. H. Heckel and R. Adams, /. Am. Chem. Soc., 1925,47, 1712.101. F-D. Boyer, P-H. Ducrot, V. Henryon, J.Soulié and J-Y. Lallemand,
Synlett, 1992, 357.102. F-D. Boyer and J-Y. Lallemand, Synlett, 1992,969.103. (a) O. Duclos, A. Duréault and J.C. Depezay, Tetrahedron Lett., 1992,
33,1059.(b) O. Duclos, M. Mondange, A. Duréault and J.C. Depezay, ibid, 8061.
104. R.G. Weiss and E.I. Snyder, J. Org. Chem., 1971,36,403.
105. (a) R.G. Weiss and E.I. Snyder, Chem. Commun., 1968,1358.(b) J.D. Slagle, T.T-S. Huang and B. Franzus, J. Org. Chem., 1981, 46,
3526.106. H.C. Brown and G. Zweifel, J. Am. Chem. Soc., 1961, 83, 2544.107. R.G. Wilt and V.P. Narutis, J. Org. Chem., 1979,44,4899.108. S. Murai, T. Murai and S. Kato, in Comprehensive Organic Synthesis, ed.
B.M. Trost, Pergamon Press, 1991, Vol. 8, part 4.4, p. 871.109. H.C. Brown, S. Ikegami and J.H. Kawakami, J. Org. Chem., 1970, 35,
3243.110. Y.D. Vankar, S. Padmashree, A. Trinada Rao and C. Trinada Rao, Synth.
Commun., 1983,13, 869.
111. A. Ookawa, H. Hiratsuka andK. Soai,Bull. Chem. Soc. Jpn., 1987, 60, 1813.
112. (a) H.B. Henbest and R.A. Wilson, Chem. Ind. (London), 1956, 659.(b) H.B. Henbest and R.A. Wilson, J. Chem. Soc., 1957,1958.
219
113. P. Chamberlain, M.L. Roberts and G.H. Whitham, J. Chem. Soc. (B), 1970, 1374.
114. K.B. Sharpless and T.R. Verhoeven, Aldrichimica Acta, 1979,12, 63.
115. A.C. Cope, A.H. Keough, P.E. Peterson, H.E. Simmons and G.W. Wood,J. Am. Chem. Soc., 1957, 79, 3900.
116. A.C. Cope, T.A. Liss and G.W. Wood, J. Am. Chem. Soc., 1957,79, 6287.117. T. Itoh, K. Jitsukawa, K. Kaneda and S. Teranishi, J. Am. Chem. Soc.,
1979,101,159.118. M. Schroder, Chem. Rev., 1980, 80,187.
119. A. San-maitin, C. Labbe, O. Monuz, M. Castillo, G. De la Fuente andA. Gonzalez, Phytochemistry, 1987, 26, 819.
120. G. De la Fuente, M. Reina, A. San Martin and J-P. Girault, Heterocycles, 1988, 27, 887.
121. H. Heaney, Aldrichimica Acta, 1993, 26, 35.122. J.R. Gillard, M.J. Newlands, J.N. Bridson, and D.J. Burnell, Can. J. Chem.,
1991, 69, 1337.123. H.C. Brown and S. Krishnamurthy, J. Am. Chem. Soc., 1972,94, 7159.124. O. Mitsunobu, in Comprehensive Organic Synthesis, ed. B.M. Trost,
Pergamon Press, 1991, Vol. 6, part 1.1, p. 1.125. W.H. Kruizinga, B. Strijtveen and R.M. Kellogg, J. Org. Chem., 1981,46,
4321.
126. Y. Torisawa, H. Okabe and S. Ikegami, Chem. Lett., 1984, 1555.127. E.J. Corey, K.C. Nicolaou, M. Shibasaki, Y. Machida and C.S. Shiner,
Tetrahedron Lett., 1975, 3183.128. O. Mitsunobo. Synthesis, 1981, 1.
129. (a) M.Barrelle, M. Apparu and C. Gey, Can. J. Chem., 1978,56, 85.(b) P. Schieber and K. Nador, Liebigs Ann. Chem., 1985, 913.
130. A.P. Marchand and R.W. Allen, Tetrahedron Lett., 1975, 67.131. D.I. Schuster, J.M. Palmer and S. Carlton Dickermann, J. Org. Chem.,
1966,31,4281.132. P. Schiess and M. Wisson, Helv. Chim. Acta., 1974,54, 980.133. A.H. Beckett, N.J. Haiper, A.D.J. Balon and T.H.E. Watts, Tetrahedron,
1959, 6,319.134. R. Noyori, Y. Baba and H. Hayakawa, J. Am. Chem. Soc., 1974, 96, 3336.135. E.J.Corey and A. Venkateswarlu, J. Am. Chem. Soc., 1972, 94, 6190.136. G. Werner and R. SchickfluG, Liebigs Ann. Chem., 729, 152, 1969.
220
137. A. Heusner and K. Ziele, Chem. Ber., 1958,91,2399.138. G.M. Sheldrlck, SHELXTL-pc, Release 4.2, Siemens Analytical X-ray
Instruments Inc., Madison, Wisconsin, 1991.
139. Vogels Textbook o f Practical Organic Chemistry, Longman press, London, 5th edn., 1989.
140. D.D. Penin and W.L.F. Armarego, Purification o f Laboratory Chemicals, Pergamon Press, Oxford, 3rd edn., 1988
141. W.C. Still, M. Kahn and A. Mitra, J. Org. Chem., 1978,43,2923.142. A.A. Qureshi and B. Sklarz, J. Chem. Soc. (C), 1966,412.143. E. Boyland and R. Nery, J. Chem. Soc. (C), 1966, 354.144. J. Thiele. Ann., 1892,271,127.
221